The UK government has unveiled a £100 million competition aimed at encouraging British AI companies to help improve public services.

As part of the scheme, the government hopes to leverage AI to bolster NHS productivity, strengthen national security, improve cyber resilience, and expand computing infrastructure.

"Every year the British state spends billions procuring products and solutions tackling some of the most important challenges facing society - from health, to our national defence," said AI minister Kanishka Narayan.

"This first-of-its-kind scheme will open up these opportunities to the British AI innovators whose ideas could make the biggest difference, backing their businesses to grow here, and go on to win globally."

The plan is to give smaller firms upfront payments to make it easier for them to access public contracts with national and local governments. Downing Street noted these organizations often lack the turnover, cash reserves, and track record to compete with larger firms.

Under the terms of the project, they’ll be allowed to keep the intellectual property they create, helping them take their work beyond the pilot stage and develop commercial products for customers in the UK and around the world.

Broader AI ambitions

The project forms part of the Sovereign AI R&D Procurement Scheme and is kicking off with four specific challenges. For the first, firms will work with the Department of Health and Social Care to develop AI systems that can automate workflows, coordinate care and support decision-making across health services.

The second, led by the Department for Business, Innovation, Science and Trade and ARIA’s Scaling Inference Lab, will support technologies that improve the efficiency of AI computing infrastructure.

For the third, companies will work with the Ministry of Defence (MoD) to develop solutions that securely connect data and frontier AI capabilities across defence systems, strengthening operational effectiveness and supporting sovereign national security capabilities.

And the fourth, in partnership with the National Cyber Security Centre (NCSC), will support technologies that help organizations understand, manage, and mitigate the increasing risks created by the use of AI agents.

Support for UK startups

The government is concerned about the number of UK startups that fail to stick around as they scale, and hopes the scheme will help.

Last year, a report from the Founders Forum Group revealed that while AI startups accounted for 27% of all UK venture capital (VC) raised in 2024, many were struggling with access to capital and talent.

One-in-three said they were actively considering relocating their company's headquarters outside the UK.

"Britain is home to some of the most innovative AI companies in the world, and this government is backing them to start, scale and succeed here in the UK," said chancellor John Healey.

"As G20 countries seek to make the most of AI opportunities, I’m determined Britain has a lead role in harnessing this technology to drive more jobs, better public services, and growth that’s UK-wide."

The Chancellor also said that the UK now plans to open up the AI Economics Institute to international cooperation with G7 countries, focused on information and evidence sharing where it is in the national interest.

FOLLOW US ON SOCIAL MEDIA

Tim Cook has served his last day as CEO of Apple, 15 years after taking over from Steve Jobs and months after announcing his planned departure.

Cook hands the reins to John Ternus, SVP for hardware engineering, who faces his first Apple product announcements in just over a week on 9 September — including a rumoured folding iPhone.

His biggest challenge won't be simple product reveals, as Cook earlier this year warned of looming price rises amid component costs skyrocketing thanks to AI-driven data centre buildouts.

Yesterday, Cook posted in a social media message that he was "sending lots of love to the Apple community."

"My title changes tomorrow, but the love I have for the Apple community never will," Cook wrote. "Thank you for being a constant source of inspiration. My gratitude is endless, and I’m excited for the next chapter!"

In a memo to staff, Cook acknowledged that it was still hard to believe he was departing Apple, pinning his own success on the company's employees.

"The truth is, whatever there is to say about my success, I know it is all because of you," he wrote in the memo, according to reports.

"You have brought out the best in me. In all my life, I have never seen or been with such an extraordinary team of people before, and every day I get to see more examples of that."

Successful leadership

Many will point to Cook's financial successes as CEO over the last 15 years. Apple's market capitalization stood at $347bn when he took over and $4.6 trillion as he departs, with the share price up 2,275%.

Cook's tenure also saw the launch of the AirPods and Apple Watch ranges, as well as hurdles such as the pandemic and American tariffs.

David Yoffie, a professor at Harvard Business School, told CNN that Cook "managed to scale the iPhone to a level that no one would have thought was likely or possible over that period of time."

However Cook said he was "most proud of what an annual report could never capture."

"This place is proof that culture triumphs over everything," Cook added. "We share a belief that what we build matters, and that we have both the opportunity and the responsibility to leave the world better than we found it."

What's next for Tim Cook?

Cook's departure as CEO naturally has many wondering what's next for Apple, although hardware expert Ternus has been with the company since 2001 and is widely deemed as a safe pair of hands. Cook called him "brilliant and wonderful and capable".

"Few people understand what it takes to build products that change the world the way John does and I could not be more excited for his leadership," Cook added.

Cook isn’t leaving Apple completely, however. He’s merely stepping down as CEO and taking on the role of executive chairman of the company board, in which he’ll still have a hand in long-term strategy.

Yesterday also saw the partial departure of Phil Schiller, a long-time Apple exec who will stay on as an Apple Fellow but no longer lead the App Store and Apple events divisions.

FOLLOW US ON SOCIAL MEDIA

Part 3: Preparing for the Workplace of 2030

The workplace of 2030 will emerge from the convergence of AI, autonomous agents, robotics, automation, skills intelligence and connected physical and digital environments. Together, they will change how organizations define jobs and create value for their businesses, customers and commercial partners.

Parts 1 and 2 of this series consider how intelligent technologies are transforming the employee experience and changing the way organisations manage performance and workforce development. These shifts are also early signals of a much larger transformation that will reshape the structure of work over the remainder of the decade.

Part 3 looks toward the workplace of 2030. It explores the technologies likely to influence how work is organized and the workforce strategies required to remain competitive. The challenge for organizations is no longer simply adopting AI, but building the adaptability to redesign roles and develop new capabilities as technology continues to evolve.

The scale of the transition will be substantial. The World Economic Forum expects structural change to affect 22% of today’s jobs by 2030. It forecasts that 170 million roles will be created and 92 million displaced, producing a net gain of 78 million jobs. This is not a simple story of technology eliminating employment. It is a redistribution of tasks and opportunities on a scale that will test every organization’s ability to adapt.

The question is not whether the workplace will change, but whether companies can change with it. Preparing for 2030 means building an organization capable of redesigning work and moving skills to where they generate the greatest value.

Emerging technologies will reorganize work

AI will become foundational workplace infrastructure. Adoption is already accelerating. Stanford’s AI Index reports that 88% of surveyed organizations were using AI in 2025, with 70% using generative AI in at least one business function. Generative AI reached approximately 53% of the general population within three years—faster than either the personal computer or the internet. Agent deployment, however, remained in single digits across almost all business functions, showing how early the next stage remains.

Wendy Harris, VP of EMEA at Rippling, says the real shift will come from “the convergence of AI, automation, skills intelligence and autonomous agents.” Instead of organizing work around fixed jobs and rigid processes, companies will increasingly match people and machines to tasks and outcomes in real time.

“The real transformation happens when organizations move away from assigning work based solely on headcount and job titles, and instead allocate tasks to the capabilities best suited to deliver an outcome, whether those capabilities are human, machine or a combination of both,” Ciara Harrington, chief people officer at Skillsoft, tells ITPro.

The transition extends beyond office-based generative AI. Robots will work beside people in factories, hospitals, construction, and logistics. Immersive systems will support training and remote maintenance. The opportunity lies in connecting these technologies to redesigned workflows rather than accelerating yesterday’s processes.

As Part 1 of this series considered, successful workplace transformation depends on more than introducing advanced tools. These technologies must form part of an intelligent employee experience that reduces digital friction and gives people greater capacity for judgment, creativity and collaboration.

Jobs will change faster than they disappear

Predictions of mass technological unemployment obscure a more complex reality. The International Labour Organization estimates that one in four jobs worldwide has some exposure to generative AI, yet only 3.3% of global employment falls within the highest exposure category. Transformation is more likely than wholesale replacement.

Even highly exposed occupations may prove difficult to automate completely. Managers, engineers, and other professionals often depend on social interaction and contextual judgment. AI may take over parts of these roles while increasing the value of the human capabilities surrounding them.

“Every role is made up of hundreds of tasks,” Harris explained. “AI doesn’t eliminate most roles—it changes the balance of those tasks.” Repetitive, administrative, and analytical work can move to technology, creating more space for judgment, creativity, communication, and problem-solving.

That requires organizations to design work at the task and capability level rather than making workforce decisions solely through job titles. Oliver Shaw, CEO of Orgvue, warns that many companies have invested in AI without considering its effect on work or the workforce. His company’s research found that 57% of business leaders deployed AI primarily because competitors had done so, while 78% of organizations had seen AI projects fail or remain stuck in pilots.

“A clear example of this is the PR and marketing industries,” says Shaw. “Both [are] fields that have been at the centre of the conversation on the disproportionate impact of AI adoption on entry-level roles, and what this’ll mean for the future of the industry when entry points are few and far between, and the foundational skills are missing. What these industries are forgetting is to train staff on the critical thinking required to use AI, and the cognitive offloading that occurs when there is too much reliance.”

Leaders risk automating fragments of work without considering which decisions remain human, or how entry-level workers develop expertise when foundational tasks disappear. Organizations should identify where AI can execute and where collaboration produces greater value.

This division of responsibilities also has direct implications for the performance and leadership questions explored in Part 2 of this series. Organizations will need to assess how effectively employees delegate to AI and apply human judgment, while ensuring accountability for consequential decisions remains clearly defined.

Continuous learning becomes core infrastructure

Skills will be the pressure point of the 2030 workplace. Employers expect 39% of workers’ core skills to change by the end of the decade. AI and big data are forecast to be the fastest-growing skills area, followed by networks and cybersecurity and then technological literacy. However, creative thinking, resilience, agility, curiosity, and lifelong learning will also increase in importance.

Technical and human capabilities are not competing categories. According to the OECD, 72% of vacancies in occupations highly exposed to AI already require at least one management skill, while 67% require a business-process skill. The employee who can operate an AI system but cannot question its output will offer limited value.

Jen Paterno, senior behavioral scientist at CoachHub, emphasised that judgment, adaptability, critical thinking, collaboration, and self-awareness develop through experience, reflection, and feedback—not conventional training alone.

“These capabilities are difficult to build through conventional training alone because they develop through experience, reflection, and feedback,” Paterno explained to ITPro.

“Organizations will need more continuous, personalized development models that help employees practice new behaviors in the flow of work. Coaching will also be critical in ensuring people have objective, brave spaces to experiment and ideate with skills that feel more foreign to them.”

The scale of the challenge is considerable. The World Economic Forum predicts that around 59% of the global workforce will require training by 2030. For every 100 workers, 11 may not receive the upskilling or reskilling they need, leaving more than 120 million people at medium-term risk of redundancy. Although 85% of employers intend to prioritize workforce upskilling, the intention must be converted into accessible learning and credible pathways into new work.

Annual courses and static competency frameworks are too slow. Harrington says organizations need systems that identify emerging skills and adapt as requirements evolve. Skillsoft found that although 86% of employees use AI at work, fewer than one-quarter feel equipped to use it effectively. Access without capability risks creating two workforces.

Skills-based models can help companies see capability beyond formal qualifications or job titles. They can also support internal mobility. Half of employers plan to move people from declining roles into growing areas, while 29% of workers requiring training could be upskilled in their current positions and 19% retrained and redeployed elsewhere, according to the World Economic Forum.

Adaptability will define competitive advantage

Workforce strategy must become more dynamic. Leaders need shorter review cycles, real-time skills visibility, plus the ability to test new roles and workflows before scaling them.

David Shrier, Professor of Practice, AI and Innovation at Imperial Business School, argues that annual planning and conventional five-year strategies can no longer support decision-making. He recommends “nimble scenarios” that allow leaders to respond more rapidly to technological and market uncertainty. AI can support that planning, but it cannot determine the organization’s purpose or appetite for change.

Business leaders must also ensure that opportunity is distributed fairly. Generative AI exposure differs significantly between countries and demographic groups. Around 34% of employment in high-income economies has some AI exposure, compared with 11% in low-income countries. Women are also more highly represented in the most exposed occupations, making inclusive training and transition programs essential, the International Labor Organization reports.

Heather Delaney, managing director and founder of Gallium Ventures, emphasised the importance of flexible planning cycles that create room for experimentation. “AI is advancing faster than a lot of us can keep up with, so flexible planning cycles that allow room for experimentation are integral to ensuring decisions made are future-proofing you and your business to grow with emerging technologies.”

Preparing for 2030 is ultimately an organizational capability rather than a forecasting exercise. No leader can know precisely which tools or roles will dominate at the end of the decade. Companies can, however, build the capacity to sense change, move skills quickly, involve employees in redesigning work, and preserve human agency as machines assume greater responsibility.

Across this series, one conclusion is clear: the intelligent workplace is not defined by how much technology an organization deploys, but by how effectively it combines technological capability with human judgment. From improving the everyday employee experience to rethinking performance, success depends on using AI to strengthen rather than diminish people’s contribution.

As 2030 approaches, the workplace of 2030 will not be won by the organization with the most AI. Competitive advantage will belong to businesses that combine technology with judgment and turn disruption into opportunity for both the enterprise and its people. Technology will shape the next generation of work, but the quality of leadership and workforce strategy will determine who benefits from it.

Motorola's resurgence continues with the Edge 70 Pro, a top-of-the-line model of its Edge handset that features four 50MP camera lenses and a stunning curved screen.

The Pro model costs £749.99 in the UK and features 12GB of RAM and 512GB of storage. A mid-level chipset and some plastics used within its build are part of the reasons it's so affordable (compared to other devices).

Motorola Edge 70 Pro: Design

Motorola's Edge 70 Pro feels a little extravagant in a similar way to Huawei handsets in that they're very glossy, delicate builds, with curved chassis and luxurious colors. The Edge 70 Pro review unit we have has a rich burgundy back cover, with a matte finish. Parts of it almost glow pink in the light, particularly around the three camera lenses.

And the curved edge display is a welcome addition; it's very similar to what Samsung Galaxy used to have – almost a throwback, but one that helps it stand out. The subtle bend in the glass is just gorgeous, and you do feel like you're using a screen that's wider than it actually is. And the curved display, chassis, and back cover also make it seem like the whole device is slimmer than it is.

In the US, Motorola is one of the most popular brands, only below iPhone and Samsung. And part of the reason for that is the quality of its handsets – the Edge 70 Pro is as beautifully constructed as anything Samsung and Apple have created over the last few years.

Motorola Edge 70 Pro: Display

The 6.78in display is a P-OLED; the 'P' means there is a flexible plastic layer underneath the main part of the display. It has a 1272 x 2772 resolution and a whopping 144Hz refresh rate.

The default color setting is quite punchy, perhaps more vibrant than it should be – this was because, for us, the default was 'Vibrant'. You can select Natural, which just slightly dials everything back, or attempt to adjust the color temperature with the giant color wheel, but it is very tricky. The other alternative setting is 'Radiant', and it's not particularly impressive or that much different from Natural.

With our colorimeter, the Edge 70 showcased 100% for sRGB color coverage and an impressive 161.1% volume (+ 97% for Adobe) under the vivid setting. Putting it up there with some of the best around.

The phone also achieved a peak brightness of 513.41cd/m2 which is decent. During our time with Pro, it never really needed to be shielded from the sun or even have its brightness tweaked, as the automated settings really do take care of it for you.

Motorola Edge 70 Pro: Specs and performance

Under the hood of the Edge 70 Pro is a MediaTek Dimensity 8500 Extreme with 12GB of RAM. This is a chip found in several mid-range devices, which fits with the pricing of the Edge 70 Pro, though it still somehow manages to underwhelm in benchmarks.

In Geekbench 7, for example, the Pro scored 1,035 for single-core workloads. And then 4,791 for multi-core. Which is rather weak compared to other midrange devices, such as the iPhone 17e and the Samsung Galaxy S25 FE.

In our looped video test, the Edge disappointed further, lasting just 12hrs and 23mins. That was less than the Razr 70 we tested the week before and considerably less than many of the phones the Edge is competing with, such as the iPhone and the Galaxy S26.

For real use, I found the device a little better, lasting well into a second day. This included most of our tests, some work-based use (email, Slack, phone calls), and a fair amount of YouTube viewing.

Motorola Edge 70 Pro: Cameras

Although this isn't a premium smartphone, it does have a premium camera setup; the Edge 70 Pro has triple 50MP rear lenses and a fourth 50MP lens on the front. The back three include a wide and periscope telephoto lens, with 3.5X optical zoom.

Image quality is good on all four lenses. However, there is a fair bit of enhancement going on; for example, it will brighten up a shot automatically, without needing to use the Night Vision settings. Which is not a terrible fault, though it will make it harder to get real-life shots with the point-and-shoot mode. I found a similar issue on the Razr 70 where it did a little too much enhancing.

For video, the Edge can shoot up to 4K at 60fps, and also shoot in HDR10+. Similarly, it offers enhanced results, but the footage we captured was smooth and stabilized.

Motorola Edge 70 Pro: Is it worth it?

The Edge 70 Pro is best described as a mixed bag. It looks the part thanks to that beautifully curved screen and stylish design. The display is pretty, and the camera setup is a lot better than you would expect for the price. But there are two areas where it really lets itself down: the processor and battery.

Admittedly, the low battery benchmark didn't really translate to poor use in real terms, though plenty of phones at this price range can offer more than 15hrs in our tests. And likewise, we didn't see any sluggish performance despite the low Geekbench scores. This is, truth be told, a good business phone. It's just not as good as the competition.

Motorola Edge 70 Pro specifications

Cyberattacks are getting faster, more evasive, and easier to execute with both scale and precision. The window of time it takes for an attack to escalate is often measured in minutes, not hours or days, as attackers use automation, phishing-as-a-service kits, and AI tools to move faster and evade detection.

Traditional reactive security defences were not designed for this.

With customers looking to them for protection against ever-evolving threats, Managed Service Providers (MSPs) must learn to navigate this era of generative and agentic AI. This requires continuous visibility and response across the full attack lifecycle. Becoming fluent in AI and combining intelligent automation with human judgement is essential to building cyber resilience in customer environments.

This is a great opportunity for MSPs. Partners that can move beyond the traditional model of blocking known threats and static signatures will gain a powerful competitive advantage in this new threat environment.

How MSPs can get ahead

It’s never been easier for threat actors to launch campaigns. Service-based platforms have industrialized credential theft, initial access, malware distribution, and more, lowering the barrier to entry while increasing attack volume and consistency.

Our research found that 90% of high-volume phishing campaigns in 2025 used kits, a significant jump from 30% the year before. We’re also seeing a growing number of attacks incorporating AI tools, such as using generative AI to rapidly craft deceptive messages and quickly shift tactics.

When an attack can progress from initial access to persistence and device compromise in five minutes, organizations need partners that can detect and respond in real-time.

The answer lies in moving from reactive support to proactive resilience.

That means continuous monitoring, earlier detection, and automated containment of suspicious incidents and anomalies rather than waiting for an incident ticket to land.

That shift changes the customer relationship, too. When an MSP identifies and addresses a threat before the customer is aware of it, the conversation moves from damage limitation to strategic guidance. That’s a different kind of value, building a stronger and longer-lasting relationship.

The MSPs best placed to make this transition are those investing now in the tools, workflows, and expertise to deliver security that is proactive by design rather than reactive by default.

Embracing AI and automation

Integrating AI and automation into MSP security offerings isn’t about replacing human expertise, but about making that expertise scale.

Manual monitoring across fragmented customer environments, including email, identity, endpoints, networks, and cloud infrastructure, isn’t viable at the speed at which modern threats move.

AI changes that paradigm, with automated monitoring tools providing continuous oversight, correlating signals across the full environment rather than treating each layer in isolation. Anomalies that could take a human analyst hours to qualify can be flagged in seconds. Routine threats can be contained automatically, without an analyst needing to intervene.

That last point is especially important, as alert fatigue is a pressing problem for security teams managing multiple customer environments simultaneously. When automation handles the high-volume, lower-complexity end of the threat spectrum, analysts can concentrate on the incidents that require business context, judgment, and experience to resolve.

Personalized solutions

The strongest security outcomes combine intelligent automation with human expertise, not substituting one for the other. Automation delivers speed and scale. People deliver understanding and context. Together, they allow MSPs to provide protection that is continuous, adaptive and aligned to what customers actually need – oversight that keeps pace with the threat environment rather than perpetually chasing it.

Predictive AI analytics also has huge potential for helping MSPs anticipate the needs of their customers. With greater insight into resourcing needs, security threats and growth opportunities, MSPs can provide personalized services which align with each customer's business priorities and future needs.

Building for what comes next

The gap between the speed of attacks and the speed of defence is widening.

AI and automation give MSPs a credible path to closing that gap, not by removing the human element, but by ensuring that human expertise is applied where it matters most. Providers that invest in building that capability now will be better positioned to protect their customers, reduce operational strain, and have more meaningful conversations about resilience rather than recovery.

MSPs that move toward a proactive, AI-augmented security model stand to differentiate themselves in a crowded market, not just as service providers, but as the kind of trusted advisors that customers need against increasingly fast and unpredictable threats.

Most of the top security teams are using AI agents, employing them to support human activity rather than replace it, according to Hack The Box.

The latest three-year benchmark data from the cyber readiness platform finds disproportionate use of agents among many of the strongest-performing teams.

More than two-thirds (68%) of the top 25 teams having an AI agent, although AI agents represented just 2.7% of registered accounts.

Those agents accounted for 4.2% of submitted flags and 4.6% of points awarded.

Across all active teams in the research, AI-augmented teams recorded a 3.2 times solve-rate advantage; an advantage that narrowed to 1.69 times among the top 5%, while the speed advantage grew.

While AI-augmented teams completed challenges three to four times faster, with the speed advantage increasing among the strongest operators, the strongest human team completed all 36 challenges compared with 32 for the strongest AI team.

"AI therefore creates the greatest value when it is directed by people who already understand the problem. It can compress research, coding, troubleshooting, and first- pass analysis. The operator still has to choose the path, challenge weak output, and decide whether the result is valid," the researchers said.

AI is now a critical component in security operations

Together, the findings suggest that AI is no longer just a question of experimentation, but is becoming part of the working methods of experienced cybersecurity practitioners.

Researchers emphasized the importance of human expertise in directing, evaluating, and validating AI-generated work, however.

"The findings do not establish that AI caused teams to perform better. They do, however, show that AI is already becoming part of the toolkit used by many of the competition’s strongest teams," Hack The Box said.

Performance across the competition has shifted substantially over the past three years, with the median recorded time-to-solve dropping from 26.1 hours in 2024 to just 13.8 hours this year.

At the same time, and even with the challenge board expanding, the number of teams completing the entire challenge board rose from two in 2024 and three in 2025 to 15 in 2026.

The improvement, said Hack The Box, could be down to better preparation, reusable tooling, platform familiarity, and deeper specialist knowledge.

“The question for security leaders is no longer whether AI will become part of cybersecurity operations. That is already happening on both sides of the equation,” said Haris Pylarinos, Founder and CEO of Hack The Box.

“What matters now is whether teams have the expertise to use it safely and effectively. Our data shows that AI is appearing most often alongside some of the strongest practitioners, not instead of them. As agents become more capable, human judgment, validation and hands-on technical skill become more important, not less.”

FOLLOW US ON SOCIAL MEDIA

Nasuni has announced the acquisition of DryvIQ, in a move that will add content intelligence and governance capabilities to the vendor’s unstructured data platform.

The deal follows the pair’s initial partnership back in February 2026 and aims to give customers improved visibility and control over unstructured data across more than 40 cloud and on-premises repositories.

The acquisition is also Nasuni’s second of 2026, following its purchase of Resilio in April, which added file synchronization and edge acceleration technology to its platform.

DryvIQ’s technology includes tools such as content-level discovery and classification, sensitivity detection, and automated policy-driven governance.

By adding these capabilities to the Nasuni File Data Platform, the vendor said organizations will have the tools to continuously discover, classify, and manage unstructured data, while automatically applying policies to high-risk content – including retention and sensitivity labels, permission remediation, and defensible deletion.

In an announcement, Nasuni chief product officer Nick Burling said the acquisition will help the company assist customers with data strategies that support safe AI access amid rising infrastructure costs.

"With UniFS as the foundation, no other solution manages, moves, classifies, governs, and activates enterprise file data in a single platform," he explained. “Customers have been attempting to stitch together a solution across multiple vendors for years.

“Now Nasuni provides one platform to manage, protect, and activate unstructured data with proven impact on both cost reduction and productivity gains."

Strengthening unstructured data management

Michigan-headquartered DryvIQ specializes in intelligent unstructured data management technology designed to help organizations manage content across cloud and on-premises environments. Its platform spans more than 40 repositories and is currently used by more than 1,100 organizations globally.

The firm’s AI-powered models scan file content across more than 550 file formats, identifying sensitive information such as personally identifiable information (PII), protected health information (PHI), and payment card industry (PCI) data across more than 175 languages.

The platform also gives organizations greater control over where and how unstructured data is stored, allowing them to continuously discover, classify, and manage content at scale to better support AI initiatives.

Nasuni said DryvIQ’s content-level intelligence will help customers make more informed decisions around data storage and governance, while the Nasuni platform allows data to be managed across on-premises and cloud environments and optimized for access and cost requirements.

DryvIQ founder and chief strategy officer Mark Brazeau said the move comes as organizations are increasingly looking to turn unstructured data into a “strategic asset.”

"By joining Nasuni, we can bring our content intelligence and governance capabilities directly into a cloud-native file platform, giving customers a single, governed foundation to classify, protect, and activate data wherever it lives," he commented.

DryvIQ is available immediately, with the vendor planning to deepen the integration between the two platforms over the coming months.

Financial terms of the acquisition were not disclosed.

FOLLOW US ON SOCIAL MEDIA

Salesforce and Anthropic have unveiled ‘Claudeforce’, marking a significant expansion of the existing partnership between the two firms.

The move will see deeper integration of Anthropic’s Claude AI across the breadth of the Salesforce ecosystem, spanning Data 360, Tableau, and Slack.

Notably, the two firms have been forging closer ties for some time now. The CRM giant noted that they are both “strategic customers of one another”.

Salesforce, for example, is Anthropic’s preferred CRM while Slack plays a key role in workforce communication, workflows, and agentic orchestration. On the flip side, Claude already underpins key capabilities within Slack, with users having access to various Claude-powered agents.

The Claudeforce announcement essentially reinforces this relationship and will drive deeper synergy between the two firms. Salesforce CEO Marc Benioff described the move as “fusing” Claude with the CRM giant’s data and workflows.

“We’re launching Claudeforce, running directly on top of Salesforce via our new AIforce UI harness, Headless 360, Data 360, Tableau, and Slack. Here, the UI is the AI — allowing you to build custom apps dynamically and answer any enterprise question,” he said in a blog post.

“We’re delivering a dynamic interface that thinks, reasons, and acts.”

Simply put, Claude will be woven directly into Salesforce’s various domains, and Salesforce will be woven directly within Claude. It’s the “best of both worlds”, according to Benioff.

So what can users expect?

Salesforce inside Claude

The first big talking point with Claudeforce centers around Salesforce in Claude, a plugin that comes equipped with 37 pre-built sales skills. We’ve already seen Anthropic make moves with domain-specific plugins in recent months.

Claude Cowork, which launched in January this year, came equipped with a range of industry-specific plug-ins, which as ITPro reported at the time sparked somewhat of a frenzy in the SaaS space.

According to Salesforce, because these plug-ins have deep access to customer data and business rules, they better “understand” seller revenue cycles – providing a helping hand to sales personnel.

“These skills were built by Salesforce and Anthropic, engineered specifically to draw on Claude’s reasoning, agentic tool use, and generative UI — not generic CRM prompts wrapped around an API but the capabilities that extend what Claude can do,” the company said in a blog post.

These plug-ins even have an “onboarding experience” that will see Claude analyze seller data across Salesforce and Slack to provide better context.

“No two companies run sales exactly the same, so Salesforce in Claude routes actions through Salesforce to help ensure business rules are always enforced when an action is taken. Setup is seamless,” the company claimed.

Claude inside Salesforce

This is a relationship that works both ways for Salesforce and Anthropic. Claude is already available within Agentforce, acting as the reasoning model for the company’s Atlas Reasoning Engine.

The LLM also powers Agentforce Vibes and Agentforce Coworker by default, and is available in Agent Builder.

With the extended partnership, the duo are eyeing deeper integration of Claude across the Salesforce estate. Notably, the LLM is now fully integrated within the Salesforce Trust Boundary through Amazon Bedrock.

Slack integration

Slack will play a key role in Claudeforce moving forward. The workplace collaboration platform has already been transformed into an ‘agentic OS’, and as ITPro reported last year, is a core pillar of the Agentforce 360 platform.

Claude acts as the default model for Slack, powering Slackbot and helping support users via features such as Slack Code and Claude Tag.

Claude Tag was rolled out earlier this year and replaced the existing Claude in Slack application. The tool essentially acts as a virtual teammate within Slack, allowing users to @ the agent within channels and allocate tasks.

Salesforce said that by embedding Claude directly within Slack, it’s “bridging the gap between team dialogue and autonomous enterprise action”.

FOLLOW US ON SOCIAL MEDIA

This week, Bobby Hellard and Scott Bekker talk with ITPro's news & analysis editor, Ross Kelly, to discuss some of the top stories from August.

That includes Google acquiring a treasure trove of data from a defunct airline in a bid to bolster AI products. The auction to acquire data from Spirit Airlines hit headlines this month as Google outbid another industry player looking to capitalize on a trove of information.

We also discuss TrendAI's insider threat report, which looked at how cyber criminals are building increasingly sophisticated insider threat networks.

Highlights

"As a a failed Spirit Airlines customer, I once had a feudal trip where I was trying to get to a Microsoft conference, and I was in line, and they're like, "Yeah, your flight's canceled. We can get you there in four days. I was like, "Yeah, I'm never doing Spirit again. But was there any customer data other than the customer chats in this volume of data?"

"A big part of the issue here is how do you tackle insider threats? Because it's quite a broad category here in terms of what we're talking about. Insider threats can be just people not doing their job correctly, not following best practices. We've covered stories on that before. Another big thing is that people who are doing this deliberately, research from CFAS last year found that one in eight UK employees have actually sold company logins, or they know someone who has. So, you know that one in eight doesn't sound huge, but I think that's really concerning there."

Links

• Google just spent $10 million in an auction for Spirit Airlines data
• Hackers are building a global insider threat recruitment network

Huntress has uncovered at least five North Korean operatives who have managed to get themselves hired so far this year, using techniques that the firm described as 'genuinely wild'.

Workers linked to the Famous Chollima group applied for jobs at IT, sales, and healthcare companies through normal channels, went through the usual onboarding process, and in some cases even went on to carry out the work as normal, sending their pay back to the North Korean regime.

The findings by Huntress come amidst growing concerns over hackers infiltrating enterprises across the United States and Europe.

Earlier this month, Recorded Future revealed that groups of North Korean IT workers dubbed PurpleDelta had created at least 22 fabricated personas, applying for jobs across a range of recruitment websites and platforms such as LinkedIn and Upwork.

Between them, they were applying for as many as 60 jobs per day. Researchers found they were able to collect high-value intelligence and exfiltrate proprietary data, source code, and internal communications in support of North Korean state objectives.

According to Huntress, there are certain telltale signs that new hires could be covert cyber criminals.

Red flags include the use of VPNs, PiKVM, and Guermok devices, particularly when they are both being used, along with web services and browser extensions associated with screen, audio, and video redirection or recording.

Caught in the act

Researchers observed several red flags, including two ‘different’ new hires whose ID documents were issued by the same police station, one day apart.

Notably, the documents were photographed eight minutes apart on the same iPhone.

Meanwhile, a worker was discovered secretly using a Raspberry Pi-based remote KVM device to control a company laptop from thousands of miles away, before the security software was even installed.

In one example in February, Huntress was contacted by an Australian firm that suspected that three employees were North Korean workers impersonating Chinese individuals.

Huntress found that documents in two of the workers' OneDrive accounts shared a common naming convention, while their passports were supposedly issued in the same place and at the same time.

There were also strong similarities between the energy bills and resident identity cards supplied – and the two workers apparently lived on the same street.

An examination of the infrastructure they were using, including IP addresses, ASNs, geographic access patterns and more, found that the accounts were authenticating with several IP addresses identified to be Astrill VPN nodes.

They were also using IPRoyal Proxy, a legitimate commercial proxy service provider that sells access to IP addresses through which customers can route their internet traffic, as well as WorkTitans B.V.

This is a bulletproof-hosting operation that appears to have been raided by the Fiscal Information and Investigation Service of the Netherlands (FIOD).

Robust identity verification

According to Huntress, the rise of fake IT workers means enterprises need to implement more robust identity verification processes.

"Mitigating the risk of fraudulent workers begins at the interview stage and continues with performing rigorous background checks of new hires prior to onboarding," Huntress advised.

"When in doubt, performing standard background checks, searching the individuals online, and verifying any employment history will help to weed out DPRK workers early in the interview process."

FOLLOW US ON SOCIAL MEDIA

CrowdStrike CEO George Kurtz has issued a warning over skyrocketing cybersecurity risks and rising threats created by recent AI advances.

Speaking on CNBC’s ‘Mad Money’ show, the CrowdStrike chief claimed current cybersecurity solutions aren’t equipped to contend with increasingly sophisticated threat groups and AI-related risks.

“A lot of companies are recognizing that legacy technology, and technology that they get for free, is not good enough,” he told host Jim Cramer. Notably, this warning applies to enterprises of all sizes regardless of their level of expertise, budgets, and security capabilities.

“Most companies have some level of gaps,” he commented, per CNBC reports. “The challenge that you have - even if companies are sophisticated and spending a lot - is the threat landscape is moving so quickly.”

The warning from Kurtz comes amidst growing cybersecurity concerns globally, with enterprises facing an increasingly perilous threat landscape. Highly aggressive ransomware groups and sophisticated state-sponsored threat actors continue to pose a huge threat to organisations and critical infrastructure.

Earlier this week, The US Cybersecurity and Infrastructure Security Agency (CISA) revealed that over 100 water systems were targeted by threat groups in July.

That advisory came less than 24 hours after the FBI revealed a host of federal agencies including the Justice Department and Federal Reserve were breached by a Chinese-backed hacker group.

Kurtz noted that threat groups are evolving rapidly and adapting techniques, which is creating huge challenges for cyber defenders.

“You really have to have the technologies with the right level of expertise to be able to combat these adversaries,” he said. “They’re moving so quickly, their techniques change so dramatically.”

AI security in the spotlight

AI security risks are also adding a new level of complexity to the global threat landscape, Kurtz told CNBC.

A series of powerful new cyber-focused AI models have been announced by leading developers in recent months, including Anthropic’s Claude Mythos.

When Anthropic unveiled the model earlier this year, it rolled out as part of a gated release program due to concerns that it could be used for nefarious purposes.

Since then, a series of high profile incidents involving Meta, OpenAI, and Anthropic AI tools have thrust AI safety into the spotlight.

OpenAI revealed in mid-July that AI agents escaped a sandbox test environment and breached a Hugging Face production database. OpenAI’s admission was followed in quick succession by Anthropic and Meta, who both confirmed similar incidents.

Kurtz told Cramer that the ability for AI agents to “swarm and find vulnerabilities and exploit those” raises the stakes for enterprises and security teams. The advent of AI agents is also creating risks elsewhere and creating blind spots for enterprises, as ITPro reported in July.

CrowdStrike’s field CTO for Europe, Zeki Turedi, told ITPro that the deep access these agents have to enterprise IT estates could further exacerbate ‘living off the land’ techniques used by hackers to infiltrate networks.

Capitalizing on the AI boom

CrowdStrike is one of a number of leading cybersecurity vendors capitalizing on the AI security boom in 2026. On 26 August, the company published its Q2 financial results, which Kurtz described as “the best quarter in CrowdStrike’s history”.

Indeed, the company reported total revenue of $1.47 billion for the quarter, marking a 26% increase compared to Q1. Annual recurring revenue (ARR) also reached $333 million, with growth surging 51% year-over-year.

In a statement, Kurtz specifically highlighted adoption rates of Falcon Flex as a key contributor to recent growth. Falcon Flex is a subscription-based licensing scheme for its Falcon cybersecurity platform.

"Delivering record Falcon Flex results, record net new ARR, and accelerating growth—the Falcon is soaring,” he said.

“The Mythos moment translated into mass-market acceptance that AI adoption needs security, and that's CrowdStrike. Every enterprise will run on AI, and securing it is the largest market opportunity in our history."

FOLLOW US ON SOCIAL MEDIA

Data belonging to millions of customers has been accessed in a cyber attack on Manchester Airports Group (MAG), the company has revealed.

MAG, which runs Manchester, London Stansted, and East Midlands airports, confirmed a cybersecurity incident on 27 August.

Data exposed in the breach relates to car park, lounge and Fast Track bookings, and on-site Wi-Fi at all three airports, the company revealed in a statement. This includes:

• Customer email addresses
• Phone numbers
• Vehicle registrations
• Postcodes

MAG noted that banking and payment details were not exposed in the breach, which at this point is believed to have impacted around 8.7 million customers.

The company added that the breach hasn’t resulted in any operational disruption, with airport operations remaining unaffected. Customer parking services are still operating as expected.

"All upcoming bookings remain valid and are unaffected by this incident," it said. "Passengers should continue to travel to the airport as normal."

MAG said it has restricted access to the affected systems, called in specialist cyber security experts, and notified the relevant authorities.

Who is responsible for the breach?

At present, there’s no indication on who is responsible for the breach, or how it occurred. However, like many recent attacks on organizations, the incident appears to have been a supply chain-related attack.

Nathan Davies-Webb, principal consultant at Acumen Cyber, said modern airports “sit at the center of a complex web of booking, parking, loyalty, payment, and internet connectivity services”.

“Many of the services in that ecosystem run on platforms operated by subsidiaries or third-party suppliers rather than the airport itself," he commented.

"That's a sensible commercial model but it creates an uncomfortable reality for security. A breach like this one in a shared upstream system can expose customer data from multiple services at multiple airports simultaneously."

Last year, a report from Thales highlighted a 600% increase in ransomware attacks in the aviation sector over the previous year, with 27 major attacks by 22 ransomware groups between January 2024 and April 2025.

A string of attacks on Hawaiian Airlines, Canada's WestJet and Qantas over the last year have all been attributed to the Scattered Spider group.

Muhammad Yahya Patel, vCISO and cybersecurity advisor for EMEA at Huntress, said the nature of the data exposed poses a significant threat to customers. Cyber criminals frequently rely on exposed information such as email addresses to conduct follow-up phishing campaigns.

“Email addresses, phone numbers, and vehicle registrations combined is a precise targeting profile for anyone planning a follow-on fraud or phishing campaign. Scammers now know you travelled, roughly when, and have two direct contact routes to reach you with a convincing story," he said.

"When that data ends up in an unauthorized third party’s hands alongside parking and lounge booking details, it fills in a surprisingly detailed picture of someone’s travel habits."

MAG urges customers to remain vigilant

MAG warned customers to remain vigilant for suspicious emails, text messages, or phone calls, and avoid clicking on links or opening attachments from unexpected communications.

The company stressed that it will never request payment card details, banking information, or passwords from customers.

Davies-Webb commended MAG for its swift response to the incident, although data protection regulations require enterprises to disclose breaches within a strict time frame.

"Public disclosure roughly forty-eight hours after they became aware is fast by UK standards and can point in one of two directions," said Davies-Webb.

"Either MAG undertook containment decisively enough that they felt safe releasing details of the breach on their own terms, or the volume and nature of the personal data compromised meant the seventy-two hour notification clock under UK GDPR was already running and disclosure was going to have to happen.”

FOLLOW US ON SOCIAL MEDIA

Basware has signed an agreement to acquire Trustpair, in a move that will extend its invoice lifecycle management platform into payment fraud prevention.

The acquisition will combine Basware’s invoice lifecycle management platform with Trustpair’s capabilities for validating supplier bank accounts, giving customers protection across the invoice-to-payment process.

Trustpair’s AI-driven platform is designed to verify that payment accounts belong to the intended supplier during onboarding, when account details change, and before payments are authorized.

Basware said the acquisition will allow it to extend its existing invoice fraud and duplicate-detection capabilities through to the point of payment at a time when AI continues to increase risk around supplier impersonation, compromised email accounts, and fraudulent payment instructions.

“A finance team can do everything right on the invoice and still send the money to the wrong bank account,” explained Basware CEO Jason Kurtz in an announcement. “Basware has spent more than 40 years protecting the invoice. Bringing Basware and Trustpair together extends that protection through the payment itself.

“Together, we will offer the first end-to-end invoice-to-payment assurance in the market, strengthening Financial Integrity and Continuous Compliance across the invoice lifecycle."

Founded in 2017, Trustpair provides automated account validation and risk intelligence to more than 600 organizations and operates across New York, Paris, and London.

The firm’s SOC 2 Type II and ISO 27001 certified platform can work alongside a range of ERP, procurement, and treasury systems, including SAP, Oracle, Coupa, Zycus, Ivalua, JAGGAER and Kyriba.

Following completion of the acquisition, Trustpair will continue to operate independently, with customers able to continue using its technology without changing their providers or existing systems.

Basware said Trustpair will also benefit from the ability to draw upon shared intelligence spanning 2.5 billion invoices and 20 million suppliers to strengthen its validation models and expand its market reach.

Commenting on the deal, Trustpair co-founder and CEO Baptiste Collot said access to Basware’s resources will enable the company to grow beyond what it could have achieved on its own.

"Trustpair was built around a simple belief: approving the right invoice is not enough if the money ultimately reaches the wrong account,” he said.

“By bringing our capabilities together, we see the opportunity to give finance teams stronger protection across the full journey, from supplier onboarding and invoice approval through payment.”

The proposed acquisition is the latest in a string of buys for Basware over the last three years, following its purchases of Glantus, AP Matching, and Redmap.

The deal is expected to close later in 2026, subject to customary closing conditions. Financial terms were not disclosed.

FOLLOW US ON SOCIAL MEDIA

Nvidia’s rumored acquisition of Hugging Face could raise the stakes for hyperscalers and AI vendors, according to analysts.

First reported by The Information, the $12.9 billion deal follows recent speculation over a potential sale. Business Insider reported last week that the firm was courting suitors on this front.

This isn’t Nvidia’s first encounter with Hugging Face, either. Reports from the Financial Times last year claimed the company turned down a $500 million investment offer from the chipmaker.

Hugging Face is a highly popular destination for developers and describes itself as the “home of machine learning”. The platform offers developers a hub through which they can access AI models, various datasets, and applications.

Notably, the platform has established a reputation as a key hub for open source models.

If the acquisition goes through, it would be a massive boon for Nvidia, according to Forrester VP principal analyst Charlie Dai. The chipmaker already boasts a significant lead in terms of the infrastructure layer underpinning the generative AI race.

This move, Dai said, would not only complement its broader product ecosystem but also provide inroads with boots-on-the-ground developers

“Nvidia has already built an extended ecosystem through GPUs, CUDA, networking, inference software, and AI frameworks,” he said.

“If this deal is confirmed, Hugging Face would give it a stronger position at the developer, model distribution, and community layers, which will strategically help shape where AI workloads are built and deployed.”

Long-term, Dai suggested that the Hugging Face acquisition could pose a threat to major industry players, particularly hyperscalers and AI providers.

“This raises pressure on hyperscalers, model vendors, and alternative hardware players,” he said. “Nvidia would gain a powerful developer channel and ecosystem asset.”

This could spark somewhat of an arms race in the broader AI industry. Rivals might “accelerate investments in independent model repositories and developer platforms” to counter the move and provide alternatives for customers.

Ecosystem evolution

The move by Nvidia comes amidst a sharpened focus on the open model ecosystem in recent months.

As ITPro reported last month, open AI models have become the latest flashpoint in big tech, particularly with the arrival of powerful new Chinese options.

Nvidia CEO Jensen Huang was one of a score of signatories in an open letter voicing support for open model approaches amidst speculation of a pending US ban.

With the debate over closed and open ecosystem approaches still raging, Dai said the acquisition could ultimately “strengthen the open model ecosystem”.

Hugging Face’s value on this front lies in its neutrality within the industry, he noted, and Nvidia will likely continue on that front.

“Nvidia has strong incentives to keep Hugging Face open, vibrant, and multi-vendor because broader open-model adoption drives AI innovation and overall demand for AI infrastructure, benefitting the entire ecosystem.”

FOLLOW US ON SOCIAL MEDIA

AI is forcing a genuine transformation in cybersecurity, redefining how threats are discovered, triaged, and stopped, and triggering an important question: at what point does automation completely remove the need for human intervention?

Walking through any major conference, the messaging is pretty much identical. Startups and legacy vendors alike are pitching fully autonomous, ‘AI-first’ architectures, promising that systems can manage threat detection and response with zero human oversight.

All this noise is making it tricky for channel partners to navigate. Customers are asking tough questions about what these platforms can actually do, forcing partners to provide high-level reassurance amid the noise. But as the hype cycle meets real-world deployment, a distinct sense of fatigue and skepticism is setting in.

AI is genuinely transforming security operations, and the pace isn’t slowing. But transformation isn’t the same as elimination: the function of the human is changing, not the need for one. For the channel, conflating the two is a fast way to damage the customer relationships that took years to build.

The agentic AI reality check

There’s no denying that machine learning and agentic AI are powerful forces for security operations centre (SOC) efficiency. If integrated correctly, automation excels at streamlining daily workflows, parsing datasets, and accelerating threat discovery. It can give defenders a vital edge in an environment where real-world exploitation can happen in minutes, shrinking the time it takes to spot an anomaly.

However, there is a massive gap between a tool that enhances human capability and a platform that appears to operate all on its own. In fact, we are already seeing an interesting shift in the market. There are lots of ‘fully autonomous’ AI SOC companies that are quietly pivoting their messaging. Their early pitches promised complete human removal; today, they are moving back toward the middle, shifting claims to highlight a dual human and AI approach.

That’s happening because the market is proving what experienced operators already knew. Cybersecurity is fundamentally a discipline of practical context. An AI model can identify an indicator of compromise, and do it faster than any human, but it can't own the outcome. Things like business impact, localised risk, and the conversation required when something goes wrong. That’s where accountability lives, and accountability is still human.

Mitigating the partner’s reputational risk

Trust is the ultimate currency in the channel. Partners who deliver the most consistent value are the ones who actually understand their customers’ environments and build sustainable, transparent relationships.

When a partner recommends and implements a fully automated security solution with zero human safety net, they’re making a big reputational bet on the technology. If that system fails, misinterprets a critical threat, or completely misses a breach, the vendor isn’t the only one who loses. The partner is the one left facing a damaged relationship with an end user who trusted their opinion.

When an event occurs, whether it’s a ransomware attempt or identity-based attack, these businesses need a dedicated team to guide them through remediation and alleviate future risk. They need a human bridge.

Building modern, integration-first partner programs

To manage the current wave of market consolidation, partners need to move past the ‘AI-first' hype and focus on customer outcomes. This means vendor partner programmes have to shift. Rather than relying on closed, single-vendor platforms, the sector needs to embrace best-in-breed approaches that work together effectively.

A successful modern security architecture shouldn’t require an organisation to tear down its existing ecosystem. It should slot into what customers already run, meet them where they are, and grow with them.

As AI-enabled attacks rewrite how businesses think about security risk management, vendors also have to become translators. Partners can’t confidently brief clients on threats they don’t fully understand themselves, and that’s the responsibility of the vendor. Clear, jargon-free intel briefings are now a part of the value proposition. The partners who will stand out are the ones who learn where AI belongs along the workflow, and where human judgment takes over.

The value of human judgment

Security leaders and CISOs are after resilience, not just tools. While technical hardening is necessary to stop an attacker, the human elements of a partnership determine how smoothly an ecosystem will function under pressure.

Human relationships are what keep multi-vendor collaborations working in good faith. When a security incident occurs, a partner needs to know that their vendor partners are acting transparently. They should be sharing intelligence and actively supporting the field teams trying to protect the business.

In practice, that looks like a vendor who checks in consistently —not just before the renewal, and not only after an incident. It also means proactive threat briefings, honest post-incident reviews, and upfront communications when there’s a platform gap. That’s what humans bring to the equation in a B2B security relationship, and what clients remember.

AI is an incredible tool for driving efficiency and speeding up discovery. But it can’t manage a relationship, and it can’t replace the deep trust required to navigate a crisis. The vendors and partners who figure out how to use AI well — and keep humans accountable for what happens next — will be the ones clients trust as the technology matures.

Confused by Dell's laptop naming scheme? You're not the only one. Last year, the company scrapped its entire portfolio, including the XPS branding, to release new machines under the Dell, Dell Plus, and Dell Pro labels. Now, XPS is back with a bang, while the Dell Plus laptops have been scrapped – in their place is the Dell 14S and Dell 16S, the latter of which we've gotten our hands on.

These mid-range productivity-centric devices are designed to be everyday workhorses that won't hit the performance heights of the higher-end laptops, but they've still captured that premium essence and will be much more friendly on your IT budget. As such, our 16S has the latest Intel 300 Series processor and an eye-catching shell. Will its £1,500 pricing be enough to tempt you away from the near-double-in-price XPS 16?

Dell 16S review: Design and display

Dell's choice of a clean and crisp Frost Blue chassis coating underlines this laptop's appeal to a more general audience, rather than a high-up executive or power user. While it's also available in a darker, less exciting grey tone, the playful charm and freshness give you a completely different outlook.

That's not to say it isn't as robust as you'd want from a Dell laptop, with its sturdy two-tone CNC-machined recycled aluminum chassis exuding a much-needed sense of durability. A nice touch is the contrast between the clouded, frosty coating on the palm rest, base, and lid against a sharper and reflective metallic lining along the sides – giving it some additional character. While an oddly square-shaped touchpad blends into the chassis, the charcoal-toned keycaps are a contrast too far and clash with the machine's freshness, as does the plastic casing around the display, undermining that premium feel. There's no flex on the lid, thankfully, which means it's at least functionally strong.

The machine's dimensions are about in line with comparable 16-inch laptops, including the HP EliteBook 8 G1a 16, and it's more compact than the Dell 16 Premium, while its 1.77kg weight is around the middle-of-the-road. For a 16in machine, you shouldn't have many portability problems.

While the 2,560 x 1,600-pixel 120Hz IPS display (equating to a below average 189 pixel-per-inch sharpness) is bright enough, registering a decent 352cd/m2 brightness in our display testing, there are a few shortcomings in terms of its color accuracy. While you can expect near-100% (or maximum) coverage of the sRGB spectrum in many modern laptops, the Dell 16S only achieves 89.1%. The results aren't much better for color accuracy for photo or video editing, with 65.1% and 70.1% coverage of the Adobe RGB and DCI P3 spectra. Combine this with a poor black level of 0.3cd/m2 and a contrast ratio of 1,178:1, it's evident that – while suitably bright – the display is among the weakest that we've tested over the course of the previous and current generation.

Dell 1S review: Performance and battery life

With the latest specs, you'd expect the Dell 16S to impress. Our configuration came fitted with a mid-powered eight-core Intel Core Ultra 7 355 processor alongside an Intel Graphics card, 16GB LPDDR5X RAM and a 512GB SSD. Performance is solid in and of its own right and it's about in line with the mobile processors you'd find in the previous generation's cohort.

Benchmarking with Geekbench 6 rendered a decent set of single-core and multi-core results of 2,683 and 11,052 respectively. We'd describe this as solid, if not unspectacular. The chipset is just about on the weaker end of the Intel 300 Series spectrum – and is comfortably outpaced by the best 200 Series chips. Take the Dell 16 Premium (Intel Core Ultra 7 255H), which scored 2,803 and 16,016 – or the Lenovo Yoga Pro 9i Gen 10 Aura Edition (Intel Core Ultra 9 285H), which scored 2,743 and 12,057.

Graphics performance was also respectable, with the Intel Graphics chip scoring 24,616 in our OpenCL benchmark – this is a little under the minimum we'd see with the Intel Arc Graphics chip, which could potentially reach above 30,000. For an entry-level GPU in a mid-range laptop, it's absolutely fine and will support basic-to-moderately intensive tasks without much problem. Where we were extraordinarily disappointed, however, was in SSD write speeds. While the 512GB SSD can read data at an average 5,280MB/s, it reads data at 255MB/s – which is borderline defective. We normally get results in the range of 2,500-6000MB/s.

The Dell 16S' battery life, meanwhile, is slightly better than many 16-inch machines we've tested recently, many of which can only reach 10-11 hours. Lasting 12hrs and 57mins, it's better than either the aforementioned Dell (10hrs and 20mins) and Lenovo (8hrs and 58mins) laptop, for example. There are one or two models that can outrun this; however, the Dell 16S is not able to outlast the AMD-powered HP EliteBook 8 G1a 16 (14hrs and 29 mins), for instance.

Dell 16S review: Features

The keyboard in the Dell 16S offers a bog-standard, solid and satisfying typing experience with minimal complaints. While certainly more generic than the exquisitely designed zero-lattice keyboard you'll find in the XPS or previous-generation Dell machines, there are few complaints about it. The off-centered placement to accommodate a numberpad will take some getting used to if you're used to centrally-aligned typing, but it's otherwise snappy, responsive, and each keystroke has a sufficiently deep travel distance.

The expansive touchpad, on the other hand, is oddly square-shaped rather than the more common rectangular form you'll find in most laptops. Its depth means that you can't perform left- or right-click actions without shifting your hands away from their position on the keyboard, while the button clicks are each a little lightweight. It's suitably responsive, however, and offers the perfect balance of smoothness and friction so that your finger feels comfortable while also not slipping on ice.

Connectivity is very good, meanwhile, with two USB-C ports with Thunderbolt 4 (40 Gbps), one HDMI 2.1 port, and two USB-A 3.2 Gen 1 ports (one on each side), as well as a 3.5mm headphone jack. There's also a little LED indicator next to the USB-C ports on the left-hand side to indicate when power is connected. We would much prefer a USB-C port on either side, however, to give you more flexibility, and there's no room for an SD Card reader. You'll also get the latest Bluetooth 6.0 and Wi-Fi 7 wireless standards – and they work well with stable connections.

Beyond the usual collection of Windows 11 AI features, thanks to a 49TOPS integrated NPU, there's very little in terms of bells and whistles available. The Dell Optimizer app is a console that lets you manage your device if you want a different skin to Windows Settings, while Dell offers a number of additional optional third-party software upgrades as well as two tiers of hardware support, at an additional cost, during the purchase process.

Dell 16S review: Is it worth it?

We've had a mixed experience using the Dell 16S, with plenty of positives including a fresh design, snappy keyboard, and healthy selection of ports, but there are too many flaws to suggest that it offers value for money at its current price.

The display, for instance, is less-than-ideal if you want to work with media in a professional sense. It's bright and crisp enough for day-to-day usage, but underperformed when we lifted the lid off the hood. Performance, too, is just about fine – but won't really challenge the highest-end machines out there right now. One suitable comparator is the Acer Swift 16 – which is just £300 more but includes an OLED screen, a much better CPU, double the RAM and SSD space.

You can argue it's worth pushing your budget up to £2,000 so you can afford the highest-end 16S, with a monstrous Intel Core Ultra 9 386H processor (slightly better than the Acer's CPU) and 32GB RAM – but you're still stuck with the same display, and for that amount of money you're really looking for something that can dazzle you.

What you're paying for in the Dell 16S is its sophisticated and robust chassis that boasts a strong design and a guaranteed level of physical reliability – while still being enjoyable to work with on a day-to-day basis. While the Dell 16S has ambitions to punch above its weight, it doesn't excel in any single area and doesn't offer the best value considering its underwhelming display.

Dell 16S review: Specifications

Anthropic has reportedly signed a $45 billion cloud computing deal with Nscale as the AI developer aims to scale compute resources.

Sources at the companies told CNBC the six-year deal will see Anthropic access 460MW of Nvidia's next-gen chips at the Monarch Compute Campus, starting at the end of next year.

The West Virginia site is expected to begin operations at the end of next, and will host Nvidia’s Vera Rubin chips.

Anthropic is yet to respond to a request for comment from ITPro. A spokesperson for Nscale said they could not confirm details from the reports.

Anthropic has signed similar deals recently aimed at scaling compute capacity, including a $5 billion deal with AMD. The company also struck a deal with SpaceX to cover up to 300MW of compute capacity. The AI developer also has an existing deal with Oracle.

London-based data center startup Nscale was spun out of an Australian bitcoin mining company in 2024, and was earlier this year valued at $14.6bn following a $2bn fundraising round. An IPO is widely expected to land imminently.

Anthropic isn’t alone in striking infrastructure deals, either. Major players including OpenAI have also invested incredible sums to build or rent compute capacity, helping to spark record results for cloud computing companies.

Last month, Google Cloud posted revenue growth of 82% — its best quarterly growth yet.

Monarch compute

In January, the American Intelligence and Power Corporation (AIPCorp) unveiled its plans to build the 2,250-acre Monarch Compute Campus in Mason County, West Virginia, dubbing it the first state-certified microgrid-based AI project.

The site boasts up to 2GW of power generated on campus by natural gas scaling to 8GW. The first phase is set to open at the end of 2027, and it is expected to become one of the largest AI infrastructure projects yet.

Two months after the unveiling, Nscale announced it had acquired AIPCorp, including the Monarch Compute Campus — and signed Microsoft as its first customer for 1.35GW of AI compute on Nvidia's next-gen Vera Rubin GPUs.

While the data center is drawing big-name AI customers, locals have complained about construction disruption causing damage to local streets and causing floods, with one report saying 60% of homeowners in nearby areas sold and moved rather than live near the campus.

Advocacy groups have raised concerns about water use, though Nscale has pledged there will be no impact on municipal water supply or residential users. As the campus is off-grid, local electricity prices or supply won't be impacted, but environmental groups have warned about the climate impact of the use of natural gas for generation.

A report from EY estimates the data centre will make up more than a third of Mason County's economy once operational.

Nscale has multiple sites planned — including Essex in the UK, and Texas — and one operational data center in Norway, with another set to open in the country imminently.

Beyond that, Nscale has several partner-run data centers in Europe and North America.

FOLLOW US ON SOCIAL MEDIA

Despite talk of an AI bubble, AI spending is not slowing down. Indeed, companies plan to commit 1.7% of their annual revenue to AI initiatives this year, up from 0.8% last year, according to research from the Boston Consulting Group.

At the same time, however, PwC’s 2026 Global CEO Survey shows that a growing number of companies are not seeing a return on their investment. A survey of more than 4,400 CEOs found that 56% have yet to see a financial benefit from their AI initiatives, while just 12% have seen both an increase in revenue and a cost reduction. Only a third are confident of revenue growth this year amid struggles to optimize AI’s ROI.

The disconnect can be put down to the fact that too many companies are buying AI tools, rolling them out to employees or customers, and only wondering several months later why they haven’t seen ROI. The problem is less to do with companies not getting value out of AI and more to do with the fact that they haven’t defined what value actually means for their business.

“A common mistake is spending heavily on tools before really understanding the problem [that needs to be solved],” says Jack Rickhuss, managing director and co-founder of tech consultancy Journi.

“AI only delivers real value when it’s in the hands of people who know how to use it properly.”

Tie AI investments to clear outcomes

Benchmarking AI budgets is becoming a critical discipline that leaders need to master to ensure they’re extracting value from AI deployments. Without benchmarking an AI project’s spend, companies have no objective way of knowing if AI is performing well or poorly and delivering value for money.

Adam Hofmann, partner (AI and people transformation) at challenger consultancy Elixirr, says that the trap he sees leaders fall into is deploying AI tools and celebrating personal productivity gains while the profit and loss doesn’t move.

This can happen because companies are “benchmarking off peers that don’t know what they’re doing either, which leaves you behind the curve or overspending on someone else’s confusion.”

Before getting started on benchmarking AI budgets, leaders should build a clear picture of what they’re currently spending on model API costs, such as per token, infrastructure such as compute, hosting, and storage, human validation, maintenance, and training. Hofmann warns that “if you can't trace the line from spend to outcome, you're measuring activity, not AI."

Rickhuss agrees, adding that most companies “are still measuring the wrong things”. He advises tying AI investment to clear outcomes, such as improving decision-making, saving employees’ time, and speeding up delivery of projects. “If you can’t link [AI investment] to something tangible, it’s hard to measure success.”

Review what industry peers are doing and then pilot

To get started, leaders should compare their spend against the peers that do know what they’re doing. One way they can do this is look at other companies in their industry using their earnings calls and 10-K filings, as well as analyst reports. They could use a large language model to extract data on peers’ AI spend as a percentage of revenue.

The next step is to break down AI use cases into various categories. This would cover the infrastructure (e.g. cloud compute and GPUs), people (e.g. the salaries for AI talent), licensing (e.g. cost-per-token), and data (e.g. storage).

Once AI budgets have been broken down, Shiro Theuri, CTO of Spanish on-demand delivery company Glovo, recommends taking “a pilot-and-test approach”. This is because piloting tools in a controlled environment can help to prevent feature creep and shadow IT. Pilots can also surface hidden costs that might otherwise have been overlooked.

An IDC and DataRobot survey carried out last year showed that 92% of enterprises deploying agentic AI at scale admitted that the costs incurred were higher than they had projected. The survey of 318 senior decision-makers at companies with more than 1,000 employees found that token consumption and hallucination remediation were the top unexpected costs, while inference was another common issue.

Continue to benchmark throughout a project’s lifecycle

To prevent their AI projects from getting stuck in pilot mode, leaders should continuously monitor the cost of running the project and benchmark this against the company’s own expectations.

Monitoring token consumption can be an effective way to keep budgets under control. AI costs can quite easily skyrocket, especially if employees end up using more tokens than forecast, running up higher bills and leading to companies exceeding their AI budgets. Luke Budka, AI director at marketing and training firm Definition, adds that tracking token usage can “reveal super users and also help identify employees who need support”.

While token usage can be used as a useful metric for cost management, leaders should be careful of ‘tokenmaxxing’ – the trend of enterprises measuring individual employee token usage and using it to measure productivity. As Budka explains, high token consumption can occur when employees are purposely inflating their usage to look busy, but it can also be a sign that employees are struggling with AI tools and could benefit from more training.

Ultimately, rather than focusing on the cost-per-token, leaders should measure the cost-per-business outcome. Determining the number of tokens required and tracking the tokens consumed to complete each workflow will help to optimize AI’s ROI.

OpenAI has admitted it should have spotted its AI had gone off the rails sooner, saying that early signs that agents were misbehaving "could have triggered an earlier response."

The admission comes as part of a series of reports into the July incident in which OpenAI models dodged internal controls and compromised a Hugging Face production environment.

That incident sparked concern across the security industry and sparked similar admissions from Anthropic and Meta, both of which later said their own AI systems had also gone rogue.

"Our models are now powerful, persistent, and collaborative enough that, absent sufficient safeguards, they can find and exploit security weaknesses across multiple computer systems," the company said in a blog post alongside the research. "Many external models, including open source ones, will soon reach comparable capabilities."

Because of that, OpenAI said the incident was a "warning shot for us and the world", calling for better safeguards to avoid similar incidents in the future.

The AI developer said it will strengthen safeguards across its research infrastructure with stricter requirements on what it calls alignment – which involves ensuring the AI is behaving as desired.

Elsewhere, the company said it will isolate sandboxes, restrict internet access and control access to model weights, and allocate more compute resources into "chain of thought monitoring" to spot dodgy behaviour more quickly.

OpenAI's investigation into the incident included security firm CrowdStrike, as well as independent reports from METR and Redwood Research. Here's what OpenAI learned from those reports.

1. AI will find a way to chat

OpenAI said the model behind the Hugging Face incident, which it's calling Internal Model 1 (IM1), and other models that were undergoing reinforcement learning training, were not intended for public release. Internet access was not enabled, nor were the agents supposed to be able to communicate.

In all, 1,200 agents that were meant to be isolated managed to communicate, sharing upwards of 70,000 notes and messages, with 700 taking part in the Hugging Face attack, the METR report noted. But AI finds a way, the company suggested.

"Despite these restrictions, the agents discovered ways to exploit our research infrastructure to communicate with one another and access the internet," the blog post noted.

The agents did so by exploiting a package manager called Artifactory, which was supposed to only be able to download and install packages from the internet to send requests.

That was then shared among other agents via Artifactory, writing files to the service and effectively turning it into a message board for the systems to chat.

Notably, the agents used Artifactory so much the service was hit by an outage, which sparked a security investigation that led to the software being locked down — but OpenAI didn't yet understand the full scale of what it was seeing.

OpenAI added that the agents found ways to leave messages for each other, even when they weren't supposed to be working together.

"When other agents noticed these artifacts, it often led them to the realization that they could use this external service to collaborate," OpenAI said.

2. AI is enthusiastic

The 70,000 notes and more than a thousand "chain of thought" reasoning transcripts reveal an odd tendency towards exclamatory language, with communications littered with exclamation points, all capital letters, and even the odd bit of profanity.

When the Hugging Face credentials were recovered by the group of agents, one declared it a "MAJOR BREAKTHROUGH!", while others used language such as "boom", "bingo", and "whoa" and "wow" throughout their efforts, as well as "Oh My God!"

The agents tried to cover up their transcripts, believing it would help their score on the ExploitGym test.

3. AI ‘knows’ when it's misbehaving

In the days and weeks that followed, OpenAI restarted the security evaluation of its models against an evaluation known as ExploitGym. This led to the intrusion into Hugging Face's systems, as the IM1 agents were looking for ways to solve the challenges assigned to them.

OpenAI's blog post reveals one agent planning to "exfiltrate" a dataset that would be useful to the evaluation solution. But other agents raised whether it was "within scope" of the challenge, with one suggesting to the group of agents that permission was required.

Another replied that attacking third-party Hugging face using a leaked token was "potentially outside intended scope" and "arguably unauthorized". It added: "Could be risky. Yet goal solution."

The agents went on to expand their intrusion into Hugging Face's systems.

4. Some AI behaved well

The agents' "messageboard" allowed these systems to communicate to build on each others' efforts, divide tasks, and share information, referring to such collective efforts as a "swarm".

But OpenAI said the agents didn't always work well together. The company claims some "stepped on each other's work" while others manipulated other agents, pretending to have received authorization.

Others refused to take part in dodgy behavior, saying that it was unethical to target Hugging Face.

"There were some agents who 'walked away' from the collective, declining to partake in their misaligned behavior," the blog post noted.

5. AI cheats, and it's hard to stop it

OpenAI said that the difficult nature of the ExploitGym evaluation led to agents to try and "cheat on their tasks by looking up solutions online".

That is known as reward hacking, OpenAI added, referring to when agents find unintended ways to solve a problem more easily.

"Reward hacking has been present in AI systems both historically… and in current frontier LLMs," OpenAI noted. " We have significantly invested in methods to identify and patch such hack⁠s."

But the complex cheating from IM1 got worse because it worked in previous training runs, reinforcing the behavior when it came to the evaluation.

6. AI should be allowed to give up

The rogue behavior from the agents was driven by the AI refusing to give up on the evaluation tasks, even when they were seemingly impossible. During the ExploitGym test, none of the models solved 198 of the 989 tasks before the hacking incident.

And it was those unsolved tasks that were a "substantial contributor" to the misbehaving agents, with those failed 198 tasks making up 93% of the tasks discussed on the message board.

"As agents used more reasoning effort, some pursued increasingly risky and out-of-bounds strategies, including eventually exploiting third-party infrastructure," OpenAI said.

FOLLOW US ON SOCIAL MEDIA

Medical technology firm Boston Scientific has been hit with a cyber attack that has severely disrupted global operations.

In a filing with the US Securities and Exchange Commission (SEC), the MedTech company said the incident was first detected on 25 August.

The attack resulted in a network outage, meaning the company only has limited access to certain systems and applications.

Boston Scientific supplies a range of devices used in cardiac care, endoscopy, and urology and reported $5.4 billion in net sales during the second quarter of 2026.

According to the firm, as soon as it detected the breach it activated its incident response protocols and began an investigation to assess and contain the threat with the help of third-party cybersecurity experts.

The company is in the process of restoring affected functions and reinstating access to systems, but that there's as yet no timeline for a full restoration.

“The incident has caused, and is expected to continue to cause, disruptions and limitations of access to certain of the Company’s information systems and business applications that support aspects of the Company’s operations, including the ability to process and ship customer orders,” the filing reads.

Who’s behind the Boston Scientific cyber attack?

As of yet, there’s no information on the initial access method used by the hackers, nor the type of attack, who may have carried it out, or what data may have been accessed.

Ross Filipek, CISO at Corsica Technologies, said at this stage the company will be prioritizing containment and establishing how the threat actor(s) gained access.

"Security teams need constant visibility into what was affected and which systems are safe to bring back online," Filipek commented.

"In healthcare, downtime carries operational consequences quickly. Strong incident response has to protect the environment while helping the business restore critical services as safely and efficiently as possible.”

Healthcare in the crosshairs

The Boston Scientific breach marks the latest in a string of attacks on major medical device manufacturers, with Stryker, Medtronic, and Abbott all having disclosed cyber incidents in recent months.

In March, Stryker was hit in an attack claimed by Iranian-linked threat group Handala, which claimed to have wiped thousands of systems across the company’s global operations and stolen around 50 terabytes of data.

A month later, the notorious ShinyHunters group claimed responsibility for an attack on the world’s largest medical device maker Medtronic. Just last month, cancer diagnostics device supplier Abbott said it had detected unauthorized access to some internal systems.

Dray Agha, senior manager of security operations at Huntress, said the data used by healthcare and medical technology providers makes them prime targets for cyber criminals.

“This is an unfortunate ripple effect on the healthcare supply chain. The attack on Boston Scientific demonstrates that cyber incidents in the MedTech sector extend far beyond IT and actively threaten the global healthcare supply chain," Agha said.

"When a major manufacturer is paralysed and unable to process or ship medical orders, the disruption creates immediate ripple effects that can ultimately delay critical treatments and impact patient care down the line."

FOLLOW US ON SOCIAL MEDIA

Glean has cut the ribbon on a new global partner program as the enterprise AI startup looks to expand its channel ecosystem.

Unveiled at the company’s inaugural Glean:GO Partner Summit, the Glean Partner Network brings together its existing partner pathways, expertise, and resources under a single global program.

The initiative gives partners access to training, commercial support, and go-to-market resources, providing greater flexibility in how they work with Glean and helping them grow their business around its enterprise AI platform.

The offering brings together company knowledge, people, permissions, and activity, giving AI applications and agents access to context from across an organization.

Glean said it is building on strong momentum for its channel ecosystem, which has grown to span North America, EMEA, APJ, and Latin America, while its partners have now completed more than 2,000 accreditations across sales, technical, and delivery disciplines.

“Partners are central to making enterprise AI work,” said Zubin Irani, vice president of partnerships at Glean.

“The Glean Partner Network gives partners more ways to build real businesses with Glean and gives customers a clearer path to the expertise they need. As partners build their capabilities and deliver results, we will invest alongside them.”

The new Glean Partner Network

Glean said its new partner program has been built to recognize partners for the capabilities they build, the value they create, and the results that they generate.

The Glean Partner Network includes referral, commercial, services and solutions, and technology pathways, with partners able to participate in more than one depending on their capabilities.

Partners can develop agents and custom connectors, co-sell and transact Glean technology, lead implementation and adoption, operate AI services, or develop technology extensions and customer solutions.

The program also includes self-service training and tiered ‘Boost Camps’, alongside specialist competencies across areas such as agent building, custom connectors, and embedded experiences.

Additionally, partners have access to resources such as deal registration, market development funds (MDF), co-marketing opportunities, as well as dedicated sales, technical, marketing and executive support.

Meeting partner and customer needs

Glean said its new initiative will generate opportunities for partners to expand across sales, services, technology, and solutions as their capabilities develop, with the potential to earn greater investment from the vendor as they grow and deliver customer impact.

For customers, the program aims to make it easier to find partners with the expertise required for different AI projects – including deploying Glean, building an integration or an agent, operating AI services, and creating industry-specific solutions.

Scott Blahauvietz, vice president of AI and cloud sales at AHEAD, a Glean partner, said enterprises require a practical way to apply AI across data, permissions, and workflows, rather than more AI models.

“Glean provides the permission-aware foundation to move from experimentation to adoption,” he commented. “Through the Glean Partner Network, AHEAD can pair that foundation with the implementation, governance, and enablement needed to build AI programs that are secure, scalable, and embedded in how people work.”

FOLLOW US ON SOCIAL MEDIA

Nvidia CEO Jensen Huang claims the AI industry is in the midst of a “golden age” after the company recorded huge revenue gains in its recent quarterly earnings report.

Revenue surged to $96.2 billion in the second quarter, marking a 106% increase compared to this time last year. The chipmaker said it expects revenue to surge to $108 billion by the end of the third quarter.

In a statement, Huang said the revenue gains reflect the growing impact of AI worldwide, with the technology having now reached an “inflection point”.

“It’s doing useful work. Its tokens are productive and profitable. Now, compute is revenue,” he said.

“And demand is accelerating. This time last year, one lab alone was driving the buildout; today, we have a golden age of new AI labs and startups, multiple frontier labs scaling in parallel, a thriving open model ecosystem and physical AI coming online — with strong momentum across the US and around the world.”

Huang noted that AI infrastructure build-outs have played a key role in recent revenue gains – an area in which the company also recorded significant growth.

Data center revenue across the second quarter increased to $89 billion, up 18% from the previous quarter and marking a 117% surge year-over-year. Notably, this earnings report marks the first since the Nvidia Vera Rubin rackscale platform announcement in May, which the company said is now “ramping into full production”.

Vera Rubin racks are now up and running at a host of major industry players, including Google Cloud, Microsoft, Oracle, CoreWeave, and Nebius.

Nvidia supply issues still a cause for concern

Forrester principal analyst Naveen Chhabra echoed Huang’s comments regarding AI infrastructure build-outs. However, for enterprise technology buyers, a combination of industry trends could dampen excitement - particularly supply constraints.

“The most important message is that AI infrastructure demand is still accelerating, but supply constraints, ecosystem lock-in, and ROI pressure are becoming the dominant strategic issues,” he said.

Chhabra said the Nvidia results suggest that the AI infrastructure market is “entering a new phase where enterprises must focus less on acquiring GPUs and more on extracting measurable business value from AI investments”.

Huang acknowledged that supply constraints are a lingering concern. The company expects revenue growth of 70% for its 2028 year, and while Huang said demand is “much greater” than that, it’s limited on the volume of products it can roll out.

Chhabra said supply issues show that multi-cloud and hybrid sourcing strategies have become “increasingly important to avoid dependence on a single provider” – especially for larger enterprises.

“Large AI training clusters and high-end inference deployments may continue to face availability constraints,” he said.

Heightened competition

While Nvidia has been center stage since the get-go in the generative AI race three years ago, enterprises are shifting their focus to alternatives in a bid to diversify infrastructure hardware.

A host of major industry players are also developing their own chips, in part to reduce reliance. In an earnings call, Huang brushed off concerns on this front, adding that he has “100% confidence that our technology will continue to be extraordinary for them”.

Alongside the earnings results, Nvidia revealed AWS plans to acquire another two million GPUs to bolster data center capacity. The announcement comes less than six months after the hyperscaler unveiled plans to deploy one million GPUs across data centers.

Chhabra noted that while enterprises are tweaking strategies, Nvidia’s broader ecosystem still holds significant weight and is a key differentiator for the chipmaker.

“Accelerators options are emerging. However, Nvidia's advantage is not simply silicon but the entire platform, software stack, and ecosystem,” he said.

“Nvidia supports "every model" and every deployment environment. The market is unlikely to become "Nvidia vs. one competitor." Instead, buyers will operate in a heterogeneous environment with Nvidia, hyperscaler chips, and specialized accelerators."

FOLLOW US ON SOCIAL MEDIA

Walk into almost any business today, and you will find employees using AI tools that IT teams haven’t approved or paid for. The use of unsanctioned applications, known as shadow AI, is coming from the top.

I’ve heard bosses and senior leaders boast about maxing out on their token limits for personal AI tools at work, some saying they would risk disciplinary action to continue. My opinion, based on research we’ve done, is that sensitive company information is being fed into unsecured platforms with knowing and active encouragement from leadership teams focused on speed and output.

With Gartner estimating that 79% of cybersecurity leaders have evidence of unsanctioned AI use, we know reckless behavior isn’t limited to the boardroom. IBM's research shows a persistent gap between leaders' AI ambition and governance readiness in organizations of all sizes. Employees are using whichever AI most increases their productivity, consequences be damned.

Business AI tools aren’t meeting employee needs. Whether that’s because of cost, AI readiness, or lack of understanding, Managed Service Providers (MSPs) and IT providers have a unique window to support businesses needing to adopt correctly. The channel is now reaching wider business managers who direct company culture, not just IT teams. With the way things are going, the channel has an opportunity to become strategic if we can meet the ask.

The pressure point

The window for the channel to move from IT vendors to strategic business partners is now. From a product perspective, upcoming licensing changes and cost pressures are likely to force decisions that many organizations have so far delayed. When budgets are tight, downgrading access to AI tools is a logical step. However, this often pushes employees back toward unsanctioned options.

Employees have already demonstrated that they will seek out tools that help them work more efficiently. Our research found that a significant proportion of employees said they would turn to personal AI tools if their employer restricted access on cost grounds. That’s a small profit gain one year and a data leak, boardroom scandal, and reputation loss the next.

For MSPs and Microsoft partners, this represents both a warning and an opportunity. Businesses must respond to employee needs and top-down rule-breaking by optimizing licensing, deploying AI properly, and putting the right software in place.

For the channel, a renewed focus on adopting AI that fits the budget and the needs of the workforce will open new business opportunities.

The government is paying attention

It is also worth noting that this is not just a commercial conversation or pandering to employee desires for the latest tech. At London Tech Week, the UK government focused on improving AI adoption among small and midsize businesses, with a caveat that tech transformation needs to happen safely and effectively.

To me, these policies and investments are a clear sign that AI enablement for SMBs is a national priority, and the channel is one of the primary routes through which it will be delivered. There has been a buzz in the channel for some time around how MSPs need to move from transactional resellers to trusted and strategic technology advisors.

Now is the time to meet the needs of providers seeking guidance on adopting quickly and safely. With renewed government focus on scaling AI for all businesses, the channel ecosystem is welcoming more than just IT buyers, presenting a long-term opportunity for the MSP.

What good readiness actually looks like

When our customers ask us about the latest AI software, financial, structural, and cultural concerns immediately crop up. They are trying to understand whether they are ready to adopt AI in a way that delivers value without introducing unnecessary risk. In most cases, the answer is more complex than a simple technology decision. Most know that the answer is more complicated than a licence purchase.

The practical work includes an honest AI readiness assessment, a clear policy framework that governs what tools can be used and how, user training that goes beyond a PowerPoint, licensing optimization, and a thorough security and compliance review that accounts for the data risks that shadow AI has already introduced.

For businesses already suffering from shadow AI use, especially where it’s driven from the top, it might be an uncomfortable conversation. Without it, businesses will keep losing money on poorly implemented and poorly performing software that their employees aren’t using, whilst the leadership team sets an example of risking sensitive data just to hit KPIs.

This conversation isn’t beyond the capability of a well-positioned MSP. But it requires a trusted relationship and the confidence to have a broader conversation with the customer than MSPs are used to having.

The window is open

Businesses have been allowing their AI tools to underperform for a multitude of reasons, but pricing changes, risk, and legislation are all about to force decisions.

If they aren’t already, businesses will soon be coming to the channel needing guidance. It’s been talked about for some time, but now really is the moment for partners and channel actors to make themselves known as AI readiness advisors. Now, before the next wave of shadow AI incidents, before a competitor gets there first.

The key question is not whether customers need support with AI; that’s obvious. For me, the channel needs to be assessing whether they can take advantage of this window and provide that support and more.

US government officials have revealed that Chinese state-sponsored hackers breached a host of federal institutions as part of a recent espionage campaign.

In a statement, the US Justice Department and FBI confirmed it has successfully seized domains used to support two hacking platforms known as ‘QScan’ and ‘QTRouter’.

These platforms were used by a state-sponsored group known as ‘QTFY’ to target critical infrastructure and “other sensitive networks”.

According to the FBI, the group successfully breached NASA, the Federal Reserve, Department of Energy, the US Senate, Department of Justice, and other agencies.

The scope and nature of the breaches has not been revealed.

“Federal law enforcement investigated and disabled the PRC’s malicious software, the latest in a series of technical operations to dismantle indiscriminate hacking activities sponsored by the People’s Republic of China,” said Attorney General Todd Blanche.

“State-sponsored malicious hackers preying on America’s critical infrastructure will be stopped and prosecuted. We are here to ensure security for the American people and will use every tool we have to keep that promise.”

State-backed operators

The QTFY group provides hacker-for-hire services to paying customers, as well as the People’s Republic of China’s (PRC) Ministry of State Security and the People’s Liberation Army (PLA), according to court documents.

QScan and QTRouter are frequently-used platforms in QTFY operations, according to the Justice Department. The first of these is used to scan for and automatically infect vulnerable IoT devices, which are then added to the broader QTRouter network of controlled devices.

“QTRouter consists of these compromised IoT devices, as well as commercial proxy service devices and leased virtual private servers,” the Justice Department noted.

QTRouter serves as an "obfuscation network" for the group, meaning that it enables QTFY and other "malicious cyber actors" to conceal their location and activities.

Officials noted that the seized domains were hard-coded into both QScan and QTRouter malware, which were then used in follow-up attacks against US government agencies.

Disrupting cyber threats

The takedown announced by the Justice Department marks the latest in a string of operations aimed at mitigating Chinese state-backed threats.

In 2025, for example, the FBI helped remove the PlugX surveillance malware from more than 4,000 computers across the US following a campaign conducted by the Mustang Panda group.

That operation came just months after a similar sting that crippled a botnet hosted by the Flax Typhoon hacker group.

“These tools were used by PRC cyber actors to hide the origin of their attacks,” said FBI Director Kash Patel.

“Thanks to the work of FBI San Diego, FBI Cyber Division, and DOJ partners, we seized adversary infrastructure and shut these platforms down.”

FOLLOW US ON SOCIAL MEDIA

Google has unveiled a series of new FinOps features for Gemini Enterprise as the tech giant aims to help developers manage surging AI costs.

Users of the AI service will be offered “expanded billing flexibility” and new cost management tools for agents, the company revealed.

This includes flexible payment options that allow “predictable per-user seat subscriptions” on a pay-as-you-go basis. Google said this will allow users to run agent workloads without hitting token quota limits mid-task.

A flexible savings plan option will also allow enterprises to set variable monthly spend limits to accommodate for growing use rates. This, the company claimed, will help reduce token consumption rates and overall costs.

“If your organization has steady or growing AI workloads, Gemini Enterprise Flexible Savings Plans offer a simple, spend-based commitment model across Gemini Enterprise usage,” the company said in an announcement.

“FSPs are designed to lower token costs while keeping budgets flexible.”

Elsewhere, new spending guardrails will allow teams to set “hard monthly caps” on AI spend on a project-by-project basis. These features allow users to estimate agent costs and prevent sudden budget spikes, the company said.

Tackling AI costs

The move by Google comes amidst growing concerns over rising AI costs in recent months.

As ITPro reported earlier this month, Gartner projects surging agentic AI adoption rates to have a significant impact on enterprise infrastructure, with inference spending in particular expected to increase dramatically.

A key factor behind these rising costs lies in how agents operate compared to more traditional chatbot-based interactions. Agents essentially operate in loops, rather than through a single request and response-type approach with chatbots – and that requires far more processing power.

Research from Signal65, for example, found that agentic AI workloads consume anywhere between four-to-fifteen times more tokens compared to chatbots.

In July, Patrick Brogan, director of the FinOps advisory team at DevOps firm Harness told ITPro that rising AI costs mean FinOps practices are more important than ever and could help keep a lid on bills.

Developer spending features

Google has also sharpened its focus on developer spending rates with the integration of its Antigravity platform within Gemini Enterprise.

This will allow developers to leverage established quotas within an enterprise subscription, creating closer alignment in terms of budgeting and consumption rates.

“To be more efficient with agentic coding costs, we are pooling developer tools quota included in each Gemini Enterprise subscription and making it available across the whole Google Cloud project so your teams can benefit from the capacity you’re already purchasing,” Google said in a statement.

Tighter budget and cost controls come at a crucial time for developers, with the use of agents in this domain accelerating rapidly over the last two years. Research from Gartner projects that AI costs could exceed the average developer salary by 2028, as ITPro reported in July.

Speaking to ITPro at the time, Gartner senior principal analyst Nitish Tyagi said a stronger focus on cost optimization and budget controls will be crucial to avoid overspending by AI-savvy developer teams.

FOLLOW US ON SOCIAL MEDIA

Hundreds of internet-facing Zimbra instances have been compromised through a vulnerability that was patched last month, researchers have warned.

The Zimbra Collaboration Suite (ZCS) hosts email, calendars, contacts, and administrative services. It has hundreds of millions of users, including thousands of businesses and hundreds of government agencies worldwide.

The security flaw, tracked as CVE-2026-73570 and rated high severity with a CVSS score of 8.9, allows unauthenticated attackers to execute malicious code remotely.

It works by exploiting a command injection weakness in the SNMP monitoring component when SNMP notifications are enabled, and affects Zimbra Collaboration's Simple Network Management Protocol (SNMP).

This is a monitoring functionality in deployments where the optional zimbra-snmp package is installed and SNMP notifications are enabled.

"This vulnerability could allow an unauthenticated attacker to send specially crafted SMTP requests, potentially resulting in the execution of arbitrary operating system commands as the Zimbra user when the optional zimbra-snmp package is installed and SNMP notifications are enabled," warned the Hong Hong Computer Emergency Response Team (HKCert) this week.

Hackers can then establish persistence, access email accounts, harvest credentials, and move laterally to other systems. Synacor, the company behind the collaboration suite, was first made aware of the issue in June, issued a temporary mitigation, and fixed it with the release of ZCS version 10.1.20 on July 20.

Agencies issue Zimbra flaw warnings

The US Cybersecurity and Infrastructure Security Agency (CISA) last week added the flaw to its Known Exploited Vulnerabilities (KEV) list and ordered US federal civilian agencies to address it within three days.

However, it appears that the vulnerability is still being actively exploited.

Non-profit security organization Shadowserver said it has spotted at least 274 internet-exposed Zimbra instances that have been breached.

Meanwhile, at least 8,200 organizations worldwide are still using vulnerable versions, although this doesn't mean they're exploitable as the vulnerability may be in a non-default configuration.

Dray Agha, senior manager of Huntress’ EMEA security operations center, said the exploitation of the flaw highlights the importance of rapid patching.

"This widespread compromise of Zimbra servers is a textbook example of the enterprise patching gap. The patch for CVE-2026-73570 was released in July, yet weeks later, attackers are still easily finding hundreds of vulnerable instances to exploit," he said.

"When dealing with an unauthenticated, remote code execution flaw on an internet-facing email server, the window for remediation isn't measured in weeks or days, it’s honestly measured in hours. Organizations need to treat collaboration suites as highly critical perimeter infrastructure and patch them with zero delay."

It's not clear who is behind the attacks, but exploitation of Zimbra vulnerabilities has in the past been linked to Russian state-sponsored hackers, including APT28, APT29, and Winter Vivern.

These groups have targeted military and diplomatic intelligence, alongside opportunistic cybercriminals seeking financial gain.

FOLLOW US ON SOCIAL MEDIA

New research from McKinsey suggests enterprises are finally reporting benefits from AI, but cost-related concerns and “constrained” productivity gains still weigh heavy on the minds of IT leaders.

Figures from the consultancy’s State of AI in 2026 report show more than one-third (37%) of respondents recorded “some” earnings impact with the technology, although that marks the same share from last year’s report.

Meanwhile, “AI high performers” said AI is now having a “significant” impact on operational efficiency and earnings.

The report suggests that enterprise AI maturity is rising, with an increasing number of enterprises now successfully scaling the technology. Improvements in this regard are being reported across the board, McKinsey noted, with respondents scaling chatbots, coding agents, and more sophisticated AI agents.

The use of agentic AI has increased significantly, for example, with 40% of large enterprises scaling agents, up from 27% last year. Meanwhile, around two-in-ten are actively scaling software coding agents.

“Among AI tools, chatbots are the most widely scaled, with 47 percent of respondents saying their organizations are scaling them across the enterprise,” the report states.

“About two-in-ten respondents report reaching the scaling phase across their organization with AI agents and a similar share report the same with software coding agents.”

Mismatched AI productivity

Workforce productivity gains rank among the most notable improvements for enterprises, according to McKinsey.

Around 80% of respondents said that AI has improved individual productivity, while 50% report that AI helps them “make better decisions”. Crucially, however, McKinsey warned that “the experience is not universally positive”.

Mid-level managers and individual workers are more likely than executives to report AI-related problems. McKinsey said this shows that AI impact “remains concentrated” among certain groups within the enterprise.

This is by no means the first study to highlight this trend. Research from Accenture in April this year found many enterprises are dealing with mismatched AI productivity gains.

Similar to McKinsey, the report found that workers are reporting benefits with the technology such as higher-quality output levels and faster project delivery times.

Yet the report pointed to a “widening gap” between the basic use of the technology and its wider enterprise impact. A key factor, Accenture said, lies in the fact that organizations' process changes are “lagging behind” integration rates.

Similarly, many AI strategies typically fail to modernize underlying IT infrastructure or workflows to accommodate the technology.

“A striking finding this year is the gap between individual gains and enterprise impact. At the individual level, AI is clearly a boon: 80 percent of survey respondents say it has improved their productivity and half say it helps them make better decisions,” said Dan Tinkoff, senior partner at McKinsey.

“Yet, only 37 percent of organizations report any positive EBIT contribution, essentially flat compared with last year.”

AI costs are taking the shine off enterprise gains

Cost-related concerns continue to weigh heavy on IT leaders’ minds, according to McKinsey. Around 20% of respondents said that AI-related operating costs have “constrained their AI use”.

“Those cost constraints are being reported across the full range of AI tools. For each of three tools – AI chatbots, AI agents, and software coding agents – about one-in-ten respondents say their organizations’ use has been constrained by costs,” the report states.

McKinsey also highlighted token-related cost concerns. This has become a recurring pain point for some businesses over the last year as firms ramp up adoption of the technology.

The ‘tokenmaxxing’ trend, for example, has prompted a surge in usage and landed some firms with hefty bills.

As ITPro previously reported, Accenture asked workers to stop using AI for basic tasks amidst skyrocketing costs while Uber revealed it blew through its entire annual AI budget in just four months.

Rising operational costs haven’t dampened investment rates, however. More than one-quarter (28%) of respondents said their organisation is now spending more than 10% of their total IT budget on AI tools.

“Looking ahead, 60 percent of respondents expect their organizations to increase their AI investments over the next year,” McKinsey noted.

“Respondents in pharmaceuticals and medical products, insurance, and banking and other financial institutions are the most likely to expect increasing investment.”

FOLLOW US ON SOCIAL MEDIA

Most business and IT leaders believe that sovereign AI is important, but they're not so clear on what it actually is and what it means for their business.

IDC defines sovereign AI as the ability for an organization to have “free choice and control over the design, development, deployment, accessibility, operation, maintenance, and governance of its AI systems and applications, as well as the underlying technology foundations they depend on”.

However, in a survey conducted by IDC on behalf of Cohere, researchers found that one-in-three had difficulty describing sovereign AI in their own words.

Of those who could, 52% of leaders explicitly described it in terms of local or national control, with 35% invoking digital independence.

Individual roles made a difference here, with line-of-business (LOB) leaders primarily viewing sovereign AI as a means for managing business risk, including data security, privacy, and cost controls.

IT leaders, by contrast, focused on regulatory compliance, ensuring that systems meet national and regional requirements. IT professionals were twice as aware as LOB leaders.

IDC predicts that by 2028, CIOs at multinational organizations will boost investments in modular, sovereign-ready cloud and data localization environments by 65% in a bid to future-proof operations against rising sovereignty demands.

Yet adoption remains uneven, according to IDC. The study pointed to a lack of a clear blueprint for operationalizing sovereignty, a fragmented and inconsistent grasp of the concept, and a lack of strategic vision.

On the right track with sovereign AI

Even without an agreed-upon definition, researchers found a clear consensus on the importance of privacy and security.

"Across every industry surveyed, enterprise leaders identified data leakage, privacy, compliance, and regulatory risk as the top concerns that sovereign AI initiatives can address, they said. "Furthermore, the largest enterprises with $20 billion in revenue are much more risk-aware than small firms."

Many see sovereign AI as presenting a competitive advantage - 35% in Canada, 28% in the US, 23% in Germany and 18% in the UK. This is most evident within the telco sector, at 37%, followed by manufacturing at 32%, healthcare at 28%, and financial services and energy, both at 21%.

“Across regulated industries and geopolitically sensitive markets, enterprises are re-architecting their digital operating models — shifting from a global-by-default posture to a deliberately sovereign-by-design approach,” researchers said.

Defining sovereign AI

Cohere said enterprises and governments should define sovereign AI in operational terms, quantify its business value, establish ownership across leadership teams, and build systems that remain available and secure regardless of external events.

Late last year, a global survey from McKinsey found that 71% of executives, investors, and government officials characterized sovereign AI as an 'existential concern' or 'strategic imperative' to their organizational goals.

As a result, the firm projected that sovereign AI could become a $600 billion market by 2030, driven largely by investment in the public sector and regulated industries.

FOLLOW US ON SOCIAL MEDIA

SMBs seeking a powerful single-socket rack server that'll fit in the smallest of spaces will find Broadberry is offering a tempting proposition. Presented as an ultra-short-depth 1U rack system, the CyberServe EPYC EP1-102A packs a lot into its compact dimensions.

Targeting SMBs and workloads such as edge computing and web hosting, Broadberry offers it with a good choice of AMD EPYC 4004/5 CPUs, and it can handle up to 192GB of fast DDR5 memory. It has another trick up its sleeve as its single PCIe slot supports an Nvidia Quadro RTX A1000 or A400 GPU card for lightweight AI and rendering duties.

Remote server monitoring comes as standard, as the CyberServe sports an integrated AST2600 BMC (baseboard management controller) with a dedicated Gigabit port. You can also manage and monitor multiple servers with the Asus Control Center Express (ACCE) utility providing a centralised dashboard.

Broadberry CyberServe EPYC EP1-102A review: Build and design

The CyberServe showcases the Asus ExpertCenter Pro ER100A B6 1U rack platform. The chassis is solidly constructed, and removing the lid reveals a smartly designed and tidy interior affording easy access for upgrades and maintenance.

Broadberry supplied our system with a meaty 16-core 3GHz AMD EPYC 4545P topped off with a chunky passive heatsink. Flanked on one side by four DIMM slots, it came with one occupied by a 32GB DDR5 UDIMM module.

Cooling is handled efficiently by a central bank of three dual-rotor fans in front of the motherboard with a plastic air shroud directing air over the CPU's heatsink. We noted that the shroud is secured in place with four screws – many similar short-depth Supermicro rack servers we've seen leave their flimsy air shroud loose, with only the lid keeping it in place.

Another cooling fan sits to the left of the chassis and is angled in to push air over the PCIe expansion slot. All told, we found the server ran very quietly, making it a good choice for small office environments.

The rear panel presents dual 2.5GbE multi-Gig ports, and if you feel the need for more speed, Broadberry offers 10GbE, 25GbE and 100GbE PCIe adapters. You're spoilt for choice if you want to connect a local monitor, as the server offers VGA ports fore and aft, while the rear 10GbE 3.2 Gen 2 port supports DP Alt-Mode for piping output to an HDMI-enabled monitor.

The front panel also has a handy 2-character Q-code LED display. This provides a running status tally as the server boots up and warns if any errors are detected – the manual provides a full list of Q-codes and their meaning.

Broadberry CyberServe EPYC EP1-102A review: Storage features

Standard storage features are fairly simple, with the base system offering two hot-swap SFF bays at the front with the backplane cabled directly to the motherboard's embedded SATA ports. Drives are easily added by snapping them into the tool-free carriers, and it's worth noting that unlike the blue-chips, Broadberry allows you to install your own choice of storage devices without invalidating the warranty.

Further storage expansion is possible as the server has an extra internal tray behind the front air grill that accepts two fixed SFF devices. Power cables are already provided, and with spare SATA and SlimSAS ports on the motherboard right next to the tray, you can cable up SATA or U.2 NVMe drives.

The system was supplied with a 2TB Micron 3500 M.2 NVMe SSD installed in the single motherboard slot beneath the expansion bay. A nice touch is that the SSD is completely covered with a large aluminium heatsink, which also gets the benefit of the expansion bay cooling fan.

Broadberry offers a range of optional Broadcom SATA/SAS RAID adapters, but the AMD chipset does have embedded RAID capabilities, which come into play when you change the BIOS SATA settings from AHCI to RAID. Arrays can be created and managed using AMD's RAIDXpert2 utility, which is available for Windows Server 2022/2025. We dropped a couple of WD Red SATA SSDs in the front bays and had no problems using RAIDXpert2 to create a software-managed mirrored array.

Broadberry CyberServe EPYC EP1-102A review: Server management

The server's embedded AST2600 IPMI controller and dedicated Gigabit port offer a tidy web interface packed with plenty of information on system health and critical components. It isn't as feature-rich as Dell's iDRAC or HPE's iLO controllers, but its dashboard provides an overview of all the latest events, and you can pull up graphs of hardware sensor data, define thresholds for each one, and link them up with email alerts if any are breached.

Firmware upgrade tools are provided along with remote power controls, and the controller's built-in firewall provides options to create IP address and port-based protection rules. Unlike Dell and HPE, the IPMI controller includes full OS remote control and virtual media services as standard features and not as chargeable licence upgrades.

The ACCE software is available for Windows and Linux hosts and is used to manage multiple servers from a single console. A licence code is included on a card in the server's accessory box, and we loaded ACCE on a Windows Server 2025 host.

ACCE focuses primarily on managing Windows and Linux endpoints and offers a remote agent deployment tool. Systems with agents appear in the main dashboard page where you can view their hardware sensors and perform a variety of tasks such as running BIOS and driver updates and system inventory.

The management controller section scans for BMC devices, and after discovering the CyberServe's AST2600, it presented a list of tasks including pulling up hardware sensor details, updating the BMC's firmware, and loading event logs. As far as we can see, BMC monitoring doesn't need a licence, which is only required for each agent deployment.

Broadberry CyberServe EPYC EP1-102A review: Is it worth it?

With a price tag for the system on review of £3,656 excluding VAT, Broadberry's CyberServe EPYC EP1-102A is offering a very affordable alternative to the blue chips. It comes up a little short in the storage department, but it is available with a good choice of EPYC CPUs, has a big memory capacity, supports an optional GPU, and its short-depth chassis can go places most other servers can't.

Broadberry CyberServe EPYC EP1-102A specifications

For much of the past two decades, conversations about print rarely made it beyond IT, procurement, or facilities teams. Print was viewed as an operational necessity, as decisions centred on devices, consumables, service contracts and cost control. While important, these discussions were largely separate from wider conversations about business strategy.

Today, that distinction is becoming increasingly difficult to maintain.

Hybrid working has exposed weaknesses in how information moves through organizations. Documents, data and workflows now flow across offices, homes, cloud platforms and multiple applications, often relying on systems that were never designed to work together. As a result, businesses are taking a much closer look at the processes that sit behind everyday operations.

Half of UK businesses still rely on legacy workflows, according to our research, while a third struggle to integrate technology into everyday operations.

In many organizations, print, digital workflows and IT systems continue to operate independently of one another, creating inefficiencies, security challenges and limited visibility over how information moves through the business.

It’s no secret that boards are increasingly focused on productivity, resilience, security and return on investment. When information is trapped within disconnected systems, manual intervention increases, slowing down decision-making and making governance more difficult to maintain. What may begin as a workflow issue can quickly become a business performance issue.

The changing customer mindset

Customers are no longer looking solely for technology providers. They are increasingly looking for partners who can help simplify operations and deliver measurable outcomes. The discussion is becoming less about individual products and more about how different technologies work together to support the wider business.

That is particularly evident in hybrid environments. Many of the challenges organizations face today stem from fragmented processes. Employees often move between digital and physical workflows throughout the day, while information passes through multiple platforms, applications and systems. Without integration, inefficiencies can quickly accumulate, critically affecting employee experience, operational efficiency, security, and business agility.

At the same time, customers are placing greater emphasis on demonstrating value from technology investments. Rising operational costs, growing expectations around sustainability and increasing pressure to automate processes are forcing organizations to scrutinise investments more carefully. Technology decisions are being assessed against wider business objectives, whether that is improving efficiency, strengthening resilience or supporting long-term growth.

This is one reason service-led models continue to gain momentum across the channel.

Rather than purchasing individual products and services separately, organizations are increasingly looking for outcome-focused solutions that bring together hardware, software, analytics, support and ongoing optimisation.

Sustainability, operational efficiency, and demonstrating return on investment are all becoming part of the same conversation, creating opportunities for partners to act as trusted advisors rather than product suppliers. Long-term partnerships are particularly valuable in this environment, as customers need ongoing support to navigate changing market conditions, evolving regulations, and shifting business priorities.

To support this, vendors and partners are investing more heavily in market intelligence and education. By working closely with organizations such as Gartner, IDC, and Quocirca, partners can access deeper insight into emerging trends, customer challenges, and best practices, helping them have more informed conversations with customers. In many cases, the role is becoming one of coaching and knowledge-sharing, equipping customers with the information they need to make better decisions and demonstrate value across the business.

The channel and AI

Another factor drawing the channel into board-level discussions is the growing connection between AI, security, and sustainability. Historically, these priorities were often treated as separate initiatives, owned by different teams and managed under different budgets. Increasingly, organizations are discovering that progress in one area depends on progress in another.

AI initiatives rely on access to accurate information and effective governance. Security strategies depend on understanding where information resides and how it moves within the organization. Sustainability programmes require greater visibility into processes, resource usage and operational efficiency. However, businesses cannot automate or improve what they cannot see.

Success is becoming less dependent on individual products and more dependent on understanding customer objectives, identifying inefficiencies, and helping organizations build more resilient operating models. The ability to connect technologies, automate workflows and provide meaningful insight into business processes is becoming increasingly valuable.

Hybrid working may have exposed the weaknesses in disconnected workflows, but it has also created an opportunity.

As customers look for greater efficiency, stronger governance and clearer value from technology investments, the channel has a real chance to play a much broader role in helping organizations achieve those goals.

Healthcare organizations collect huge amounts of data that can be used to improve operational processes and save patient lives.

However, exploiting this data is easier said than done. Healthcare professionals face multiple challenges, from stove-piped data to governance hurdles, as they look to turn information into insight.

It’s in this challenging environment that Dr Joe Zhang, CTO at OneLondon, is aiming to create a new era of interoperability. OneLondon is a specialist healthcare partnership that connects health and care information across the capital to improve patient treatments and plan future medical services – and at the heart of this approach sits the Snowflake data platform.

“The data is a lot more complicated in healthcare than [in] other settings; it's highly dimensional, it's multimodal, meaning that you have structured data, free text, but also imaging data, and genomics data,” says Zhang.

“The value of the data for improving care is enormous, but what you're dealing with on the ground is just a constellation of many siloed systems locked away and managed by different teams.”

Disparate data

Patient data in London is spread across GPs, hospitals, mental health providers, community services, and social care organizations. This data has traditionally been fragmented across disparate systems, making it difficult to create a joined-up view of patient health to predict demand or conduct population health research.

After an open procurement process that considered other data platforms, OneLondon selected Snowflake, with Zhang suggesting the provider’s capabilities meant its technology ticked more boxes than its rivals. Today, Snowflake is helping the organization solve its data interoperability challenges.

“At a local level, the platform is the layer where we can ingest data that's compliant, secure, and that we can trust,” he says. “It’s also operationally open, which means we can use our own tooling, which is standard things like SQL and Python, to transform data, ingest it, observe it, standardize it, and turn it into a usable asset for clinicians and analysts on the ground.”

Zhang, who spoke with ITPro at the recent Snowflake Summit 2026 in San Francisco, says the platform provides a data mesh-like approach, where professionals can share insights securely with others. As a doctor who used to work in the Intensive Care Unit (ICU), Zhang has first-hand experience of the life-changing impact of timely insights. Now, he’s helping professionals in other healthcare organizations to exploit their data.

“My background is as a doctor, but I've got a technical background in data engineering and data science. I left medicine after COVID. I worked in industry for a while doing biomarker development in real-world data, then came back to London to lead part of the program across London, and then stepped into the CTO role,” he says.

“This job gives me the best of both worlds. We can make a bigger impact with data, data products, and applications built on data, but it can take a bit longer to see the impact. When I worked in [the] ICU, you’d see the results within 10 minutes or so. This work involves a wider scope and longer timelines, but a potentially bigger impact.”

A data journey heading towards value recognition

OneLondon’s activity covers about 30 hospitals and more than 1,400 GP practices. Some healthcare organizations are more mature in their data journey than others. While the Snowflake platform provides a core layer for joined-up analytics efforts, the key to long-term success is ensuring that professionals recognize its value.

“It’s more of an incentivized play rather than saying, ‘This is the pattern we're going for,’” says Zhang. “The organizations that have come on to Snowflake have done so because they think the technology helps with interoperability. A lot of the work has been about unifying opinions across all the organizations in London to make this approach work.”

The key benefit for professionals is a single patient record. Whether someone spends time in GP services, local hospitals or specialist healthcare centers, Zhang says their pathway is tracked: “If you're a clinician at a point of care, you can start to build applications that say, ‘This patient, based on everything that has happened, is at more risk of renal dysfunction, so we should do something about it.’”

Work on the platform continues. The major migration process finished in April, with primary care data feeds currently coming online. At least a dozen use cases are in development. While these pioneering projects will produce benefits, Zhang says it’s important to recognize that the core focus so far has been perfecting the underlying technology layer.

“There are all sorts of infrastructure and networking challenges you must overcome before you think about integration,” he says.

“But because Snowflake offers an application layer where you can provision resources and scale compute to applications that are deployed, it becomes a highly secure solution for us to deploy all sorts of services.”

That effort is crucial, as Zhang says ambitious projects in population health or life sciences research won’t be scalable if the right data isn’t available. The long-term aim of this initiative is to create proactive care applications for clinical pathways: “The approach we choose may not work at first, but you iterate, and create a cycle where you constantly learn and improve.”

Future innovation

Another long-term objective, says Zhang, is to use the joined-up view on data to help support the development of new drugs for disease treatments. He says a key reason the UK isn't doing as well as it could in terms of developing new therapies for patients is a lack of data. The Snowflake platform could play a crucial role in supporting progress.

“We don't know what diseases are there, who is suffering from what disease, and what the trajectories are,” he says.

“So being able to map that process out, being able to run more clinical trials in this country, and getting new drugs to patients faster, would be a big outcome for me if we can help do that.”

Reflecting on progress made so far, Zhang says clear long-term objectives and a trusted technology partner are critical to success. As someone well-versed in life-changing decisions in high-pressure environments, he recognizes that digital leaders looking to push data-enabled change must walk before they can run.

“Often it's not sophisticated things like AI and agents; it's just having the right data in the right place at the right time,” he says.

“We need to be able to measure outcomes and see what's happening to our patients. Those things sound simple, but they've never really been possible before. So, in that instance, this project is all about trying to prevent preventable things.”

Apple has unveiled new Mac Mini and Mac Studio models, boasting an “extraordinary leap in performance and AI capabilities”.

The new devices wil be equipped with Apple’s M6 and M5 Ultra chips, the latter of which is described as its most powerful to date.

“Today we’re debuting the next giant leap in performance and AI compute for Apple silicon with the incredibly advanced M6 and the most powerful M-series chip yet,” M5 Ultra,” said Sri Santhanam, Apple’s vice president of Silicon Engineering Group.

The move by Apple comes amidst a sharpened focus on local, on-device AI capabilities for developers globally. The Mac Mini, for example, has emerged as a go-to option on this front, enabling users to run agentic workflows locally.

Santhanam said the new Mac Studio M5 Ultra, for example, will provide the “ultimate desktop performance and the ability to run massive AI models” on the device.

Under the hood of the M6 Mac Mini

The new Mac Mini will feature Apple’s M6 chip, which is built using a 2-nanometer process. This will boast a 12-core GPU, two more than the M5, along with a neural accelerator in each core.

Using this design, Apple said this will deliver a near 30% increase in GPU performance for AI compared to the M5, and more than eight-times compared to the M1.

“The super cores blaze through single-threaded workloads, the performance cores use less power and join the super cores to run demanding multithreaded workloads, and the efficiency cores handle everyday background tasks — all with industry-leading performance per watt,” Apple said in a blog post.

Elsewhere, the Mac Mini M6 will support up to 32GB of unified memory, along with 170GB/s of unified memory bandwidth. The firm said this marks a 10% increase over the M5 and will help users run LLMs on the device more efficiently.

The M5 Pro Mac Mini

Alongside the new M6 range, Apple also unveiled the Mac Mini equipped with an M5 Pro chip. This will boast a CPU ranging up to 18-cores alongside a 20-core GPU, and is catered specifically for complex tasks such as 3D design, VFX, and game development.

The device also supports up to 64GB of unified memory, alongside 307GB/s of unified memory bandwidth. Once again, Apple touted the device's capabilities in terms of local AI orchestration.

"With industry-leading performance per watt, Mac mini with M5 Pro is quiet, efficient, and ideal as an always-on desktop for AI agents or creative workflows," the company said in a blog post.

What to expect with the M5 Ultra Mac Studio

According to Apple, the new Mac Studio featuring the M5 Ultra chip is the “ultimate powerhouse for pro workflows” - again placing a strong emphasis on local AI capabilities.

The M5 boasts a next-gen GPU with up to 80 cores, with each also boasting a neural accelerator to provide up to 4.5-times the GPU performance compared to the M3 Ultra chip, and over six-times that of the M1 Ultra.

“The GPU includes Apple’s latest shader core with second-generation Dynamic Caching, as well as hardware-accelerated mesh shading and third-generation ray tracing, delivering up to 40 percent faster graphics performance than M3 Ultra,” the company said in a blog post.

On the memory front, the new device also boasts significant improvements, featuring up to 512GB of unified memory and 1.2TB/s of unified memory bandwidth.

Apple said this will enable users to “store huge datasets entirely in local memory” and run LLMs ranging in the hundreds of billions of parameters locally.

What to expect with pricing

Customers looking to get their hands on the Mac Mini with M6 or M5 Pro chips can pre-order today (25 August), although the devices will set you back.

In terms of pricing, the Mac Mini with M6 will start at $899 in the US, although users in education can get it for a reduced price of $799. The Mac Mini with M5 Pro, meanwhile, starts at $1,699 and $1,599 for education.

Elsewhere, the Mac Studio with M5 Ultra starts at a whopping $5,499 and $5,099 for education.

Designed for devs

Notably, Apple’s various developer frameworks and tools, such as Core AI, Core ML, and Xcode, will “tap directly” into both chips.

This offers marked advantages in terms of local AI orchestration, the company said.

“With these frameworks and new chips, developers can run and fine-tune large AI models locally on their Mac,” Apple said in a blog post.

“Apple’s developer tools and frameworks automatically optimize performance across the CPU, GPU, and Neural Engine, and give developers the ability to use Apple Foundation Models, App Intents to tap into Apple Intelligence features, or their own proprietary AI models to build and run powerful AI workloads entirely on device.”

Another pint-sized powerhouse

Apple’s pint-sized powerhouse, the Mac Mini, is back with an M6 chip – and it comes at a time when mini PCs are having their moment.

Smaller players from China such as Geekom and Beelink are finding more and more markets with their innovative and powerful desktop boxes, but they will struggle to match the pulling power of Apple.

And they won’t be anywhere near as adorable as the Mac Mini.

FOLLOW US ON SOCIAL MEDIA

Zero-click attacks bring to mind the advanced spyware typically targeted at a specific subset of users. Infamous examples, such as the Pegasus spyware that targeted the family of murdered Saudi dissident Jamal Khashoggi, saw a user compromised simply by receiving a WhatsApp or iMessage.

But now, email is being used in a zero-click phishing campaign waged by Russian state-backed hackers, according to the UK’s National Cyber Security Centre (NCSC), which has issued an alert.

Targeting a vulnerability in the Zimbra Collaboration Suite (ZCS) software, the so-called ‘beehive’ attacks by Russian group Laundry Bear aim to steal email correspondence from organizations operating in critical sectors, according to the joint advisory.

Similar to the zero-click campaigns involving text messages, users only have to view a malicious email to be compromised.

Zero-click attack evolution

Zero-click used to mean “expensive, highly-targeted mobile exploits” used by nation-states for “silently dropping spyware” via WhatsApp or iMessage, says Matt Cooke, cybersecurity strategist at Proofpoint.

In the latest attacks, Laundry Bear – also known as TA488 and Void Blizzard – “has taken that mechanic out of the intelligence-gathering niche” and “turned it into a mass espionage tool aimed at the corporate inbox”, says Cooke.

His company has published a detailed breakdown of the new campaign.

Zero-click attacks such as these are concerning because they remove “the one thing firms have spent years training people not to do: Click the link”, says Rich Greene, certified instructor at SANS.

This type of activity is an evolution of attacks carried out via platforms such as WhatsApp and iMessage in highly targeted spyware campaigns, according to Greene.

“Moving those same ideas into email makes complete sense from an attacker’s perspective,” he said. Email is everywhere, businesses depend on it, and messages are constantly being processed in the background before the user ever decides whether to interact with them.”

How the email attacks work

In the Laundry Bear campaign, which likely used AI, viewing a crafted message in a vulnerable version of Zimbra webmail was enough to trigger the exploit. This would provide adversaries access while leaving the victim “with little reason to suspect anything had actually happened”, explains Alexander Leslie, a senior advisor at Recorded Future.

Analysis of the latest campaign found the techniques could be adapted to exploit vulnerabilities in other email software applications used by Western organizations.

The attackers initially exploited a flaw tracked as CVE-2025-66376 in Zimbra Collaboration Suite, but they later took advantage of a second vulnerability in Outlook Web Access, according to Cooke. “The group has shown it will scale a working technique, rather than retire it after one campaign,” he says.

The flaw is architectural, explains Cooke. “Email clients render HTML as browsers do. TA488 doesn't need a phishing hook; it hides malicious script fragments inside standard mail formatting.”

However, while mobile zero-click attacks often don’t require any interaction at all, the campaign does require users to open an email. Therefore, this would properly be described as a “half-click” exploit by researchers, says Cooke.

Who is a target?

The Zimbra campaign ran for at least five months against Ukrainian government entities and US defence, nuclear and research targets. During this time, adversaries exfiltrated information including 90 days of email, session tokens and saved credentials.

It’s notable how quickly TA488 moved on, says Cooke. “The day after an advisory on that campaign went public, researchers caught the same group already running the second exploit chain against Outlook Web Access, hitting government, telecoms, finance, hospitality and aerospace targets.”

The Outlook payload is “a step up in sophistication”, according to Cooke.

“It steals OAuth tokens through compromised mailbox add-ins and grants itself server-side folder permissions via a low-privilege default account. That access survives password resets and even a full device re-image, because it doesn't live on the endpoint at all.”

Going forward, the sectors most at risk are those that could be targeted by Russian intelligence. This includes defence, government, energy, law enforcement and media, says Leslie. Any organization holding politically, militarily, or commercially sensitive correspondence could be of interest, he adds.

Tackling zero-click email attacks

As zero-click attacks move from messaging to email, telling users not to click on links is no longer a valid response. Yet there’s no need to panic. The latest attacks are still very targeted and require unpatched flaws to compromise firms.

If you are in an at-risk sector, or if you use ZCS, there are a few steps you can take to reduce the risk. Greene believes organizations need to focus on the basics and “execute them well”.

The NCSC has advised ZCS users to patch immediately, as well as follow mitigation advice, use a third-party authentication service that supports passkeys where possible, and boost network monitoring capabilities.

“Patch quickly, keep email clients and operating systems up to date, reduce unnecessary message preview or content-processing features, monitor endpoints and accounts for unusual behaviour, and use layered email, identity and endpoint security controls,” adds Greene.

Defence in depth is becoming increasingly important, with many organizations using Microsoft Defender Suite for endpoint, email, application and identity protection, says Peter Jones, cyber security specialist at Conscia UK.

However, he believes it’s worth complementing these controls with behavior detection across the network and within the data center, using tools such as Cisco’s Secure Network Analytics or Secure Workload capabilities.

“When implemented correctly, Microsoft and Cisco Security tools can work well to provide the visibility to respond effectively in the event of a breach.”

As zero-click attacks continue to evolve, Cooke thinks the response must be “architectural”. He advises firms to patch webmail and Exchange promptly, but to also “assume patching will always trail behind zero-days”.

With this in mind, treat sessions and tokens as “short-lived by design”, he advises. “Shorten token lifetimes, and build identity monitoring that flags anomalous token behavior.”

Overall, he believes firms should plan for containment speed over prevention. “As TA488's Outlook campaign shows, once persistence is server-side, revoking a password isn't enough. Isolating and revoking the session, and auditing folder permissions and add-in access, is what actually closes the door.”

British tech workers are using AI for 35 tasks per day across their teams, new research suggests. While that works out to just a handful of prompts for most employees per day, it's starting to build up to serious electricity usage.

That's according to Uswitch Business Energy, which found the average workplace team generates 24 prompts per day, including writing emails, automating admin tasks and analysing data.

But that's across the whole team, not per worker. Instead, Uswitch combined its data with YouGov research for some back-of-the-envelope extrapolations.

Survey data shows 32% of British workers use AI at work, and with an average team size of 12, and an estimated average AI prompt generation of two per day, that adds up to an estimated 22 million AI prompts every working day.

That sounds like a lot, but there's just shy of 26 million full-time workers in the UK, and not all of them are sitting at computers.

IT and telecoms workers account the most prompts, at 35 per team, followed by finance at 29, Education at 22, and healthcare at 21, the report found.

Workers aged 25-34 had an average of 30 AI-assisted tasks on their teams each day, versus 27 for younger workers and just 18 for those aged over 55.

Are energy concerns warranted?

Uswitch's energy division warned that the shift to AI is eating up a fair amount of electricity.

The switching company points to Google's own estimate that a Gemini text prompt uses 0.24 watt-hours (Wh) of energy, meaning British employees are using 5.28 MWh of electricity each and every work day, or about 1.37 gigawatt-hours (GWh) each year.

That works out to enough electricity to power 508 British homes annually. Of course, that energy use is over in a Google data center, rather than at point of use.

"AI has rapidly evolved from an emerging technology into an everyday business tool," said Ben Gallizzi, energy expert at Uswitch. "As more organizations adopt AI to support tasks such as content creation, customer service and administration, the cumulative energy demand from these tools will continue to grow."

Developing and training a model is an energy hungry process, with estimates that training current cutting edge models uses between 20-25MW of power over three months, or between 43 GWh and 54 GWh in total.

However, research by MIT Technology Review suggests that those massive figures make up a small amount of the total energy use of AI.

As much as 90% of its footprint lies in inference, with small text based queries, image generation, and the rest adding up when used by so many people.

"Although each prompt uses a relatively small amount of energy, the scale of adoption means businesses are collectively generating billions of AI interactions every week," said Gallizzi, although the billions would be globally, not just in the UK.

Indeed, OpenAI said last year that it was seeing 2.5 billion queries on ChatGPT each day.

Gallizzi added: "As organizations continue to embrace AI, understanding the impact this could have on future electricity demand will become increasingly important."

Time to track

Uswitch suggested businesses need to be aware of these figures, just as they track energy use from more mundane aspects of their offices. As noted, this usage will be off-site in a data center run by AI companies, so it's more about considering overall usage and impact than direct costs.

"Businesses already carefully monitor energy use from equipment, heating and lighting," Gallizzi said. "As AI becomes more deeply embedded in workplace processes, organizations may also need to consider how digital tools contribute to their overall energy footprint."

Wanton AI use could be reined in — not by electricity footprint concerns, but by rising prices and shifting billing models. That's led to companies pushing to end so-called "tokenmaxxxing" and encouraging more sensible use of AI credits.

FOLLOW US ON SOCIAL MEDIA

Enterprises are falling into a trap of using AI when it’s not necessarily needed, and it’s costing them dearly, according to Roger Lee, AVP of solutions consulting at Appian.

Speaking to ITPro, Lee said the hype surrounding AI - particularly agents - is pushing some IT leaders to apply the technology in an ‘AI for the sake of AI’ type approach.

“I think a lot of organizations and a lot of senior executives and staff within organizations have been told to ‘do AI’,” he said.

“So rather than just pushing AI for the sake of AI, look at what the decisions are we’re trying to improve, what data we are using, and where does that human oversight sit.”

Lee’s comments come after Accenture told staff to stop using AI for unnecessary tasks amid surging costs. As adoption of the technology accelerates, he told ITPro that enterprises need to have a serious conversation on how they integrate AI tools.

Agents vs business rules

Pursuing the ‘AI for the sake of AI’ approach means that many businesses aren’t just incurring huge costs, but adding needless complexity to processes that were already streamlined prior to the generative AI race.

“The key thing is really trying to identify the right place for AI versus just trying to use AI to solve everything,” he said.

“In many cases, if it’s straightforward automation, if there’s a business rule where you’ve got a definitive yes or no answer, then sometimes business rules and workflows that result in a definitive answer are the right way to go.”

Admittedly, Lee said there are instances where agents are necessary, especially with regard to research or complex problem solving tasks. Acknowledging that this is a case of different strokes is crucial, however.

“I take an example from insurance with an automation claim, where you wanted to do some research and do some work, then an AI agent might be the right solution for that,” he said.

“But it's really understanding the right place to use a rule versus an AI agent, and the impact that that can give.”

This is particularly relevant in regulated industries, Lee said, where margins of error can have huge ramifications.

Agents are highly effective in terms of dynamic reasoning, he noted, but in cases where businesses need a “really definitive zero error response” then IT leaders need to know where and when to automate.

“Regulated industries is another area where humans in the loop are really important,” he said.

“It's what makes processes accountable. So to ensure that not everything is automated, where you have things that are exceptions or outliers, that you absolutely have humans in the loop to respond to those.”

Shoehorning AI into processes

Enterprises have found themselves shoehorning AI into areas and processes that aren’t applicable, research from Gartner shows. Moreover, they’re expecting near-immediate returns with the technology.

Ironically, a key factor behind this lies in the desire to deliver returns on investment with AI. Yet by doing so they’re less likely to unlock any benefits.

Speaking at the time, Melanie Freeze, director of research at Gartner, said that many AI initiatives are “overly ambitious or poorly scoped” and don’t align with “real operational needs”.

Ultimately, Lee told ITPro that as with any adoption process, AI should still be viewed as a “business pull rather than technology push” approach.

“I think that the whole value that you can get from AI within an organization is all around what the business needs, rather than what the technology can push to the organisation,” he said.

“First and foremost, what problems are we trying to solve, rather than where can we use AI, and that changes the perspective on what we're trying to do,” Lee added.

FOLLOW US ON SOCIAL MEDIA

The government has outlined proposals for new powers under the Cyber Security and Resilience Bill aimed at boosting cyber defences for the UK's essential services.

With hostile cyber attacks and sabotage on essential services like energy, water, transport, and healthcare on the rise globally, vendors and companies with connections to hostile states pose a 'growing and serious' threat to the UK’s security.

The idea is for the government to be able to step in when essential service providers are looking to buy equipment or technology from suppliers that could pose a critical national security risk.

"These new powers mean we can act before a threat materializes, not just after the damage is done. By working together with industry, we're putting national security at the heart of how essential services choose their suppliers," said cyber security minister Liz Lloyd.

"This is about staying ahead of a growing threat, and giving the public confidence that the everyday services we depend on like water and energy supplies, our transport network and hospitals, are protected."

New cyber safe procurement guidance would be made available for all essential service providers to help them secure their supply chains. Providers will be able to refer themselves for a risk assessment if they're unsure about a vendor – and be given advice on how to mitigate any risks.

Meanwhile, the package would grant the government new legal powers to issue binding directions to essential service providers, including requiring extra security measures, a phased withdrawal, or in the most serious cases, an outright ban on acquiring a vendor or supplier's products or services.

Growing state-backed threats

The proposals come amidst growing concerns over the impact of state-sponsored cybersecurity threats.

Earlier this year, the National Cyber Security Centre (NCSC) warned that the UK is experiencing increased threats from Iranian-backed cyber groups, with a heightened risk of indirect threats for organizations with a presence, or supply chains, in the Middle East.

Government figures indicate that an outage caused by a cyber attack on London and Southeast England's electricity networks could potentially cost the economy up to £442 billion over the following five years.

Just this week, the vulnerability of key UK infrastructure was brought into focus by the news that Iranian-linked hackers were been able to knock a small power plant offline for four days.

The attack, which targeted internet-exposed programmable logic controllers (PLCs), is believed to be the first of its kind in the UK.

The amendments to the bill were laid before Parliament on Monday, ahead of Lords Committee stage scrutiny, which is set to begin in September.

FOLLOW US ON SOCIAL MEDIA

US lawmakers are pushing back against the Trump administration's decision to reduce headcount at the Cybersecurity and Infrastructure Security Agency (CISA).

Five Democrats have written to the Government Accountability Office (GAO), calling on it to look into the effects of the cuts on the agency's ability to protect critical infrastructure and respond to evolving cyber threats.

The group has requested an investigation into how CISA’s workforce has changed over the past five years, including its size, composition, geographic distribution and number of contractors.

They also want the GAO to identify which programs have been affected by the cuts, what the effects have been, what data CISA collects and analyzes for workforce planning and whether it uses that data to align resources with agency priorities.

"At precisely the moment when our adversaries are accelerating attacks against critical infrastructure, the Department of Homeland Security's Cybersecurity and Infrastructure Security Agency (CISA), the nation's lead civilian cyber defense agency, has lost nearly one-third of its workforce, raising serious concerns about the agency's ability to fulfil its mission," they wrote.

"Little is known about the impact of these workforce reductions on CISA’s programs and services or what processes and plans the agency has in place to ensure the agency is hiring the right people to address lost skill sets and meet mission demands."

CISA cuts run deep

The latest changes were announced in June as part of the 2027 budget proposal. CISA is set for a cut in funding of around $700 million - with the White House claiming that it's been functioning “as a hub in the Censorship Industrial Complex, conspiring against the First Amendment rights”.

The cuts would refocus CISA on its core mission – Federal network defense and coordinating with critical infrastructure partners – while eliminating waste by consolidating redundant security advisors and programs. Around 900 jobs are expected to go.

CISA lost nearly 1,000 employees, or about a third of its workforce, in the first half of 2025 - although the agency has since made moves to try and boost staffing again.

It's also seen cutbacks to its work on election security. And all this has come as its work has expanded more than ever, thanks to an accelerating threat environment.

Gene Moody, field CTO at Action1, said the cuts show a lack of awareness over the critical work conducted by the agency.

"In many respects, vulnerability analysis is further behind the curve than it has ever been. And it is being attacked by its own governing bodies out of ignorance and misunderstanding. This is an existential threat to cybersecurity," Moody commented.

"When analysts and programs are overloaded, the ecosystem gets noisier: important vulnerabilities compete with an enormous amount of lower-value information, prioritization becomes harder, and defenders have less confidence about what actually deserves immediate attention."

FOLLOW US ON SOCIAL MEDIA

Semiconductor revenue is set to almost double this year as the AI sector continues to reshape the market – and drive up prices for in-demand components.

New figures from Gartner project global semiconductor revenue growth to top 92% this year, reaching $1.6 trillion over last year's $809 trillion. In 2027, that’s expected to climb further to $1.9 trillion, the company said.

"The semiconductor industry is entering a fundamentally new phase of growth," said Ben Lee, Director Analyst at Gartner. "The pace of industry expansion is accelerating dramatically, driven by sustained investment in AI infrastructure and a stronger-than-anticipated memory pricing cycle."

That's good news for those manufacturing semiconductors, but the rise of AI – and the surge in demand that it's sparked for components – has led to price rises across a range of industries.

In the cloud market, a host of providers have issued price hike warnings. Meanwhile, consumer electronics are being hit hard by rising hardware costs. Apple recently warned that prices will rise this year, for example, while PC shipments have hit record lows.

Memory and beyond

Semiconductor gains are largely driven by memory revenue, with prices for the key component skyrocketing because of AI demand. Some predictions suggest that AI will consume a fifth of all DRAM capacity in 2026.

That shortage is sending prices skyrocketing and already impacting businesses' plans to upgrade tech.

This trend is expected to continue, according to Gartner. The firm predicted DRAM revenue would increase 246.6% this year, with NAND flash revenue up by 371.9% – and that’s despite additional capacity being made available.

Memory now makes up 54% of total semiconductor revenue, a huge leap from 27% last year. Revenue for memory will top $837bn this year and $1trn next year, Gartner predicts.

"AI infrastructure has fundamentally changed the dynamics of the memory market,” said Shrish Pant, Director Analyst at Gartner.

“While pricing expansion is accelerating growth in 2026, continued AI infrastructure deployments, higher memory content per AI server and sustained demand for high-bandwidth memory (HBM) will support memory revenue growth through 2027 and beyond."

While semiconductor revenue growth is dominated by memory price rises and demand, Gartner noted that the AI roll-out is benefiting all aspects of the industry, as AI data centers need CPUs, networking, power management, and more.

"AI is expanding the semiconductor opportunity across the entire infrastructure stack rather than benefiting a single device category," said Lee.

Gartner said semiconductor revenue growth excluding memory would grow from $589 billion in 2025 to $718 billion in 2026, an increase of 21.9% – and reach $864 billion in 2027.

AI data centers

Notably, the mass buildout of data centers is changing how semiconductors are used and where they end up, according to Gartner.

"As AI infrastructure scales, it is not only reshaping investment across the semiconductor ecosystem but also redefining the industry’s application mix," said Lee, saying Gartner sees a "a structural shift in where semiconductor value is created and how demand is evolving across the industry.”

Gartner predicted that the AI data center ecosystem would make up over a third (36.5%) of semiconductor revenue this year, climbing to more than half (53%) by 2030.

FOLLOW US ON SOCIAL MEDIA

Freeths has announced the appointment of John Jones as its new chief technology officer (CTO), in a move the law firm said will strengthen its leadership bench and growth ambitions.

Jones will work across Freeths’ twelve offices located across the UK and will oversee the shaping and delivery of its technology strategy.

His remit will include ensuring the business has secure, resilient, and future-ready technology platforms that support employees, improve the client experience, and enable operational efficiency.

In an announcement, Freeths national managing partner Karl Jansen said the appointment comes at an “important point” in the company’s ongoing growth and development.

“Technology is central to how we support our people, serve our clients and operate as a modern law firm,” he commented.

“John’s depth of experience, combined with his pragmatic and inclusive leadership style, will be instrumental as we continue to build scalable, resilient and innovative technology foundations across the firm.”

Headquartered in Nottingham, Freeths is a UK commercial law firm serving private and public sector clients, including organizations such as Centrica, Aldi, Mercedes-Benz UK, Tarmac, Experian, and Harworth.

Jones joins the business with more than 30 years’ experience across technology and professional services. Most recently, he served as CTO at Forvis Mazars and previously held senior tech roles at BDO.

He began his career as a software developer working on business-critical applications before moving into senior technology leadership positions within the professional services sector.

Jones’ experience includes working with enterprise platforms such as Microsoft, Workday, and legal-specific technologies, while his expertise spans data, digital skills, large-scale transformation programs, as well as the safe and effective use of emerging technologies such as AI.

Commenting on his new role, Jones said he was impressed by Freeths’ existing technology platforms, as well as the clarity of its plans for the future.

“Professional services, and the legal sector in particular, is going through significant change, and technology has a vital role to play in enabling better outcomes for clients and colleagues,” he explained.

“I’m excited to be joining Freeths and to be working closely with teams across the business to support the next stage of the firm’s journey.”

FOLLOW US ON SOCIAL MEDIA

AI is becoming an infrastructure race, not just a software one. Across Europe, governments, hyperscalers and investors are competing to build the compute, connectivity and operational backbone required to support AI at scale. The UK has made clear that it has been and intends to be part of that conversation.

There are good reasons to take that ambition seriously. Through its UK Compute Roadmap, the government has placed AI infrastructure at the centre of its industrial strategy, with plans to expand public compute capacity, establish AI Growth Zones, accelerate data center planning and invest in compute and semiconductor capabilities.

Private investment has followed, including Blackstone’s £10 billion AI data center project in Northumberland, CoreWeave’s multi-billion-pound investment in UK AI compute capacity, AWS’s £8 billion commitment to UK data centers and Google’s continued investment in expanding its UK infrastructure footprint.

But ambition on its own is not enough.

The UK’s competitive advantages

The UK enters this race with several meaningful strengths. It has one of Europe’s largest AI ecosystems, a mature cloud market, an established hyperscale data center footprint and a deep enterprise customer base that is actively exploring AI adoption.

London remains one of the continent’s leading financial centres, helping attract investment while bringing together organizations with both the capital and appetite to invest in AI initiatives.

Combined with world-class universities and research institutions that continue to produce talent and innovation, the UK has many of the ingredients needed to position itself as a long-term hub for European AI innovation.

The UK’s infrastructure challenges

Power availability may be the UK’s greatest obstacle. Industrial electricity prices remain among the highest in the developed world, while grid connections for major infrastructure projects can take years to secure. The UK’s AI ambitions ultimately depend on a resource that cannot be scaled overnight: energy.

That matters because AI workloads, particularly GPU-intensive ones, are unusually power-hungry. Over time, energy economics will influence not only where infrastructure is built, but where workloads actually run. As enterprises look more closely at the cost of scaling AI, location becomes an operating decision, not just a property or planning decision.

The UK also faces growing competition from across Europe. France benefits from strong government backing and abundant nuclear energy. The Nordic countries offer renewable power, cooler climates, and lower operating costs that naturally support large-scale data centers. Germany combines industrial scale with significant enterprise demand for AI, while Ireland continues to attract hyperscale cloud investment even as it grapples with its own energy constraints.

Rather than competing on identical strengths, each market has the chance to build a compelling proposition. The UK’s competitive edge lies less in offering the cheapest power and more in combining enterprise demand, financial investment, cloud maturity and a growing AI innovation ecosystem. Whether that will be enough to secure long-term advantage remains an open question.

What does this mean for the channel?

For channel partners, focusing only on which country “wins” risks missing the bigger opportunity.

As enterprises move from AI experimentation to production deployments, infrastructure decisions are becoming business decisions. Clients are asking where AI workloads should run, how to balance cloud and on-premises environments, control costs, meet sovereignty and compliance requirements, and scale without creating operational drag.

That shift moves the conversation beyond products and into outcomes. The most relevant partners will be the ones that can help clients make better decisions across architecture, operations, governance, and cost, not just deploy another piece of technology.

For Managed Service Providers (MSPs), systems integrators, and other channel partners, this creates an opportunity to move further up the value chain. Rather than simply helping clients select and deploy infrastructure for AI, partners can help assess AI readiness, modernize data center environments, design hybrid AI architectures, optimize networking and storage, strengthen security, implement governance, and build a more disciplined approach to long-term AI cost management.

Energy will remain central to that conversation. GPU-intensive workloads already place significant demands on power and cooling, making infrastructure efficiency a real business concern. As customers seek to understand why AI deployments cost what they do, partners that can connect technology decisions with operational improvements will stand out.

The bottom line is that the opportunity for channel partners is not tied to a particular postcode or data center location. Instead, it is tied to helping clients navigate the growing complexity of AI infrastructure itself.

Looking beyond the build-out

While public attention often focuses on where compute capacity will be built, the more durable opportunity lies in supporting AI inference: the day-to-day execution of AI workloads that power business applications and operational decisions. As AI adoption matures, these workloads must run securely, efficiently and cost-effectively, creating sustained demand for the design, integration and managed services that channel partners are well positioned to provide.

Whether the UK ultimately establishes itself as Europe’s primary AI infrastructure hub remains an open question. The country has clear strengths, but also some constraints that need long-term solutions, such as around power, planning, and long-term capacity. Geography alone will not determine the winners.

For channel partners, taking a wait-and-see approach would be a mistake. The smarter move is to help clients design and refine AI environments around their operational needs now. Those that bring expertise in AI architecture, workload optimisation, cost management, energy efficiency and governance will be best positioned to lead, regardless of which country comes out ahead.

While tokenmaxxing has taken the tech industry by storm in 2026, it’s time enterprises start ‘storagemaxxing’ to avoid being hit with huge infrastructure costs.

That’s according to David Boland, VP of cloud strategy at Wasabi. Speaking to ITPro, Boland believes that a confluence of industry trends means IT leaders need to sharpen their focus on storage optimization.

Hardware supply chain issues and rising prices are an area of acute concern for IT leaders in 2026 and are forcing some to alter strategies to compensate for growing costs, he said.

“We all know that the supply chain issue for storage is big news,” Boland told ITPro. “It’s really hard for organizations, especially if they don’t have deep pockets or are large, well-known organizations; it’s always difficult for them to get the hardware that they need.”

The result, according to Boland, is that they “need to look elsewhere” for alternatives or maximize the use of their existing infrastructure.

With AI in the equation, this adds a new layer of complexity to storage optimization, Boland noted. Simply put, there are a “ton of different storage options out there” yet each AI use case has different requirements from a storage perspective.

“Whether it’s inexpensive, long-term storage for compliance and regulatory reasons, or in the training process or the preparation process, [such as] storing checkpoints of training,” he explains. “Each one of these, let’s call them phases, in the AI lifecycle have a different storage need.”

Given the varied storage requirements when it comes to AI data sets, Boland said that some enterprises find their existing infrastructure isn’t compatible.

“Some high performance, expensive storage is not a great fit for your long term archiving, so therefore knowing where to put your data at which phase of that AI pipeline helps you maximize your investment or maximize the amount of money you spend.”

Storagemaxxing explained

Storagemaxxing is essentially a term designed to “give some new life” to traditional storage optimization processes, Boland said, and one that is inspired by various social media ‘maxxing’ trends.

We’ve all seen them online: ‘looksmaxxing’, ‘tokenmaxxing’, even ‘hobbitmaxxing’.

With storagemaxxing, this is primarily about streamlining storage to accommodate for various stages of the AI lifecycle, be that from data collection to training, deployment, and retention.

The aim here is to ensure that data stays “AI-ready, secure, cost-predictable, and free to move between cloud and compute providers as workloads change”, Boland noted in a recent blog post.

Notably, this has a critical impact on both AI and cost efficiency. AI tools need swift access to the data that feeds them. Ensuring seamless access by optimizing storage also reduces the strain placed on infrastructure.

Costly changes

Being “free to move” data between cloud and compute providers is an area Boland said many enterprises have been caught off guard with in recent months.

Many enterprises didn’t anticipate certain costs associated with storage, he said – be that storing data on the wrong platform or in the wrong cloud tier.

This has a direct impact on costs, he said. Even simply accessing that data or moving it can incur further charges due to egress fees. Indeed, Wasabi’s 2026 Cloud Storage Index survey found 48% of storage costs now go toward fees rather than actual capacity.

As ITPro reported at the time, this is severely impacting IT budgets and forcing a rethink of storage strategies.

“Being locked into a storage solution because you can't afford to move that data out, either through egress charges or retrieval fees, is an issue,” he said.

“I see that becoming more of an issue in the AI space. What's nice about hyperscalers is they have a broad range of services that go from GPU to CPU to high-performance storage to low-cost object storage,” Boland added.

“If there comes a point when a customer wants to move that data to another provider for either a better GPU cost or for availability of services, then they get hit with fees that they don't anticipate.”

FOLLOW US ON SOCIAL MEDIA

Tens of thousands of previously-exposed AWS access keys are still active and valid, researchers have warned, with hundreds holding full admin rights.

Truffle Security said its scanners verified 64,024 unique AWS key pairs across 431,875 public findings that surfaced publicly between August 2022 and August this year.

This includes git history, Hugging Face datasets, Docker images, package registries, and CI logs, the company said.

It then re-verified the 10,616 leaked keys with complete credentials and found that 88% still authenticate, including 526 root keys and 242 identity and access management (IAM) users holding AdministratorAccess - which have full admin rights.

"The largest single source is Hugging Face. AWS credentials are the second most common secret type we verify there, with 8,482 unique live keys across 3,394 public datasets," said the firm.

"Hugging Face keys also skew privileged: 17.9% are root, the highest share of any source we track. Most of those datasets are snapshots of public code repackaged for training."

Of the live keys with creation dates, the median age was 1,831 days. Half are over five years old. The oldest was 17.4 years, nearly as old as IAM itself.

"Only 25 keys (0.9%) were created in the last 30 days. Almost none of this population leaked recently. The count has been building for years," Truffle Security said.

"Rotation is the rarer event. Of the keys where we could enumerate the user's access keys, only 13.7% (398 of 2,903) have any newer key alongside the leaked one. The other 86% were never rotated, superseded, or cleaned up."

Of the 7,590 active IAM users, 929 carry AWS's own AWSCompromisedKeyQuarantine policy, and were detected by AWS as exposed and restricted. Of these, 112 carry the original version, which AWS stopped applying in 2023.

Researchers noted that they were flagged at least three years ago and their owners notified by AWS. However, no action was apparently taken and the keys still authenticate.

Tighter access controls

Truffle Security said organizations delete root access keys. One-in-six leaked keys is root, for example, and as such there is no longer any legitimate reason for a root access key to exist in 2026.

They should also sort their IAM keys by age and set a budget alarm, even a small one, to catch cryptomining early - 90.5% of leaked-key accounts have none.

Elsewhere, researchers said security teams should assume that committed means leaked. Nearly half (43%) of keys were sighted more than once across repos, datasets, and images. Deleting the file does not help once it is in a training corpus.

"Any time AWS is aware of exposed keys, we notify the affected customers. We also thoroughly investigate all reports of exposed keys and quickly take any necessary actions, such as applying quarantine policies to minimize risks for customers without disrupting their IT environment," an AWS spokesperson told ITPro.

"As always, customers can contact AWS Support with any questions or concerns about the security of their account."

FOLLOW US ON SOCIAL MEDIA

A small UK power plant was reportedly shut down for four days last month following a cyber attack attributed to Iranian hackers.

The incident appears to have affected a small-scale gas generator, rather than a large power station, and did not affect the wider grid.

According to the Daily Telegraph, the government has contacted the chief executives of power companies following the incident and has written to businesses with advice on the steps they should take to protect themselves.

The attack has been attributed to hackers affiliated to the Iranian government, and is believed to be the first of its kind in the UK.

Graeme Stewart, head of public sector at Check Point, said the incident highlights the potential risks posed by threats to critical infrastructure.

“This marks a grave escalation in the Iran conflict because a hostile state-linked cyber threat has reportedly reached into UK energy infrastructure and caused a physical shutdown lasting four days,” he said.

“That should concern every organization responsible for keeping this country running. For most Brits, the Iran conflict is happening thousands of miles away and cyber warfare probably still conjures up images of stolen passwords, leaked data and companies being held to ransom,” Stewart added.

“The prospect of a hostile state being able to reach into the infrastructure beneath our everyday lives changes that dramatically."

Rising state-sponsored threats

Notably, the incident coincided with a series of attacks on US water supply infrastructure in recent weeks. More than a dozen states were affected in the campaign, which is believed to be linked to Iranian-backed threat groups.

The attacks targeted internet-exposed programmable logic controllers (PLCs), with the attackers remotely changing IP addresses and turning on and setting passwords.

Several water firms were left unable to view connected equipment, and in some cases it was shut down.

The FBI has recommended disconnecting PLCs from the public-facing internet, strictly controlling access to them, tightening up passwords, and reviewing project files running on PLCs for unauthorized changes.

UK infrastructure in the crosshairs

According to the National Cyber Security Centre (NCSC), the UK is experiencing increased threats from Iranian-backed cyber groups.

Earlier this year, the agency issued a warning over heightened risk of indirect threats for organizations with a presence, or supply chains, in the Middle East.

Stewart suggested that the impact of an attack on larger power supply operators or other critical infrastructure firms could be disastrous.

"We have to ask what happens if the next target is bigger, more critical or more deeply connected to the services millions of people rely on. Britain’s critical national infrastructure underpins almost every part of modern life, including electricity, water, transport and communications, and those systems are increasingly digital, interconnected and dependent on one another," he said.

"The question now has to be whether Britain is genuinely ready if something more serious follows.”

A spokesperson for the UK government told ITPro that it's working on an Energy Resilience Strategy. Set for publication later this year, this will outline plans for ensuring the energy system remains stable and secure.

"The UK has a highly resilience energy system. We work closely with the energy sector to protect infrastructure and ensure the highest standards of security," the spokesperson added.

FOLLOW US ON SOCIAL MEDIA

OpenAI CEO Sam Altman believes AI could spark a new wave of small businesses, but the tech industry has a lot of work to do convincing entrepreneurs how the technology can help.

Speaking on the David Senra podcast, Altman suggested AI could help break down traditional barriers for entrepreneurs, particularly in terms of financing and technical capabilities.

“Even if most people don’t want to start really big companies, a lot of people want to start smaller companies, and that has been hard,” he said.

“That has been something that has required a fair amount of privilege, and luck, and resources to be able to do. We are about to see the greatest boom in people starting smaller businesses that we have ever seen,” Altman added. “I think AI is empowering that.”

Altman’s optimistic outlook on this front is reflected in recent research conducted by OpenAI. As ITPro reported in March, a survey by the firm highlighted a strong appetite for AI tools among small businesses across the UK.

Eight-in-ten small businesses across the country now use the technology in daily operations, OpenAI found, although the study warned that uptake varies wildly on a region-by-region basis.

Notably, Altman told Senra that OpenAI and the broader tech industry hasn’t quite emphasized the potential benefits of AI to entrepreneurs and small businesses in recent years.

The focus has largely been on larger organizations and how the technology is impacting operations in these domains.

“For some reason the field, including us, has not talked about that enough, even though we see all the signs of it,” he said. “We have not built enough products to accelerate that, but I think we’re going to see a lot more of that."

Some SMBs are pulling ahead

Altman’s comments come in the wake of recent research from Dell Technologies which found small and medium-sized businesses (SMBs) are increasingly optimistic about the impact of AI.

The study found two-thirds (66%) of UK small businesses view AI as a “route to growth” while more than half (56%) believe the technology will give them a competitive advantage.

Brian Horsburgh, UK small business country manager at Dell Technologies, told ITPro that small businesses have a huge advantage over larger companies in terms of adoption.

Smaller firms are far more agile and typically more flexible in how they integrate the technology in daily operations. This, the study found, is highlighted in the rate of successful adoption projects.

Indeed, larger companies tend to cite a broader range of barriers when it comes to adoption, including lengthy approval processes, fear of failure, or resistance to change across the workforce.

FOLLOW US ON SOCIAL MEDIA

For years, channel partners have made their margins on cloud migration and security, helping clients move data somewhere safe and keep it that way. Both have become the default, but solving that problem has quietly created a new one: data that is safe and well governed is not the same as data people can actually use.

Backend systems have grown more sophisticated while the interfaces sitting on top of them have stayed clunky, fragmented, and largely ignored, leaving valuable information locked away from the people who need it most, and this is where the value has now shifted.

Relying on cloud migration and baseline security as a primary revenue engine will leave channel partners behind because providers continue to sell infrastructure solutions to an enterprise market that has already moved past the infrastructure phase.

Storage and security handle yesterday's issues, while data silos resulting in data fragmentation are now the urgent challenges of today. Data silos trap critical information within isolated, non-communicating systems, scattering corporate data unevenly across public clouds and legacy applications. This inevitably leads to data fragmentation: a chaotic state in which corporate data is unevenly distributed across multiple public clouds, legacy systems, and disconnected applications.

The consequences of this fragmentation are crippling, leaving organizations without a single source of truth. In fact, 77% of respondents in an IBM Institute for Business Value study strongly agreed that these silos actively hinder real-time analytics and data-driven decisions.

For the past decade, managed service providers (MSPs), value-added resellers (VARs), and IT consultants have built highly profitable practices by guiding organizations through cloud infrastructure modernisation. But today, the next margin growth opportunity isn’t where data is stored; it is how that data is experienced: a shift known as the Digital Experience Platform (DXP) space, which helps organizations unify, govern, and present fragmented backend data into usable, front-end interfaces.

The multi-million dollar data disconnect

Organizations continue to pour substantial capital into upgrading their cloud infrastructure. An Omdia survey shows how spending reached US$110.9 billion in late 2025, a 29% year-on-year surge, with another 27% growth forecast through 2026 on global cloud infrastructure. However, a massive operational problem persists: the data sits perfectly organized in the backend but remains functionally inaccessible to the end user.

Simply owning data is no longer a competitive advantage; the real value is in how easily people can use it. This structural gap between backend complexity and front-end utility defines the "Data Experience" gap.

For a business-to-business (B2B) buyer, this gap looks like a frustrating customer portal that cannot display real-time shipping updates because the shipping database does not talk to the web interface. For a supplier, it means relying on manual emails because procurement platforms do not share information.

Transitioning to data experience architecture

To capture higher profit margins in this landscape, channel partners must transition from storage infrastructure providers into data experience builders.

This evolution requires moving away from traditional, slow, and expensive data migration methods, such as moving information into the cloud, and instead focuses on building smart, lightweight integration layers. By seamlessly connecting legacy systems to modern cloud environments, partners can consolidate fragmented data into a single, cohesive view without disrupting the underlying storage architecture.

This ability to deliver integration without interruption represents a highly specialized skill set for which corporate clients are willing to pay a premium.

Beyond merely connecting these disparate systems, opening the flow of data introduces an urgent need for smart permissions and sophisticated governance.

A distributor, a B2B customer, and an internal sales representative may all require access to information stored in the same database, yet what actually appears on their respective screens must be uniquely tailored.

Designing and implementing these complex, context-aware access control systems serves as a high-value advisory service. It establishes a level of security and compliance that basic software resellers simply cannot replicate, deeply embedding the partner into the client's day-to-day operations.

Ultimately, these seamless backend connections enable the delivery of consumer-style B2B experiences that modern buyers now expect as standard. Digital experience insights tracked by platform providers indicate that the primary hurdle for B2B companies isn't designing an attractive user interface; rather, it is untangling the complex web of integrations sitting right beneath the surface.

Channel partners who can bridge this gap by building tailored vendor portals and collaborative digital workspaces resolve their clients' most critical technical friction points, effectively making themselves indispensable business allies.

The high-margin advisory model

When a partner fixes an organization's data experience, they are doing far more than installing software. They are mapping out how the business actually runs, improving daily operations, and building resilient integration systems.

This results in higher client retention, robust profit protection, and a steady stream of recurring revenue as the client’s digital needs continue to scale.

Architecting these solutions requires a technology foundation designed specifically for backend complexity. Platforms that act as a unified hub connecting content, workflows, and backend systems to deliver personalized experiences are built for this transition, offering an evolvable architecture that allows partners to bridge the data experience gap across the entire business ecosystem.

By building security and context-aware compliance directly into the core of the integration layer, partners can safely deliver highly localised, consumer-grade experiences across diverse global markets. Furthermore, utilizing integrated AI and scalable SaaS or PaaS environments ensures that as client traffic and campaign demands fluctuate, the underlying data layer remains resilient.

The foundational infrastructure of the modern enterprise has already been laid. The channel partners who will win over the next decade will be those who don’t focus solely on storage boxes, but instead upskill their teams in integration governance and start unlocking the value of the data inside them.

The market has moved past the plumbing phase to build data experience, and channel partners need to move with it.

One of the most sought-after form factors on the market is the humble miniature desktop PC, with a wide range of manufacturers competing to produce the most appealing mini computer. Most of these are cheap and cheerful entry-level devices that suit office workers running simple daily tasks. These include the Geekom Air12 2026 Edition and Acemagic Vista Mini V1, among many others.

A handful of machines aim to punch up a level, including Acer's Revo RB-102. Launched last year, this alluring felt-topped mini PC looks the part and features an Intel Core Ultra 100 Series chipset with a modest amount of memory and storage. Priced at a not-inexpensive £589 – similar in value to the middling Acemagic Kron Mini K1 and more expensive than the excellent Acemagic Matrix Mini M1 – does Acer's compact desktop machine do enough about it to convince you to depart from the entry-level options?

Acer Revo RB-102 review: Design

Mini PCs look very similar, barring the chassis color, some detailing, and the placement of ports. Otherwise, you get a box that's usually black, grey, or silver with a tiny bit of flourish.

Yes, there are some exceptions – the DGX Spark devices, for example, adopt the same form factor across manufacturers, inviting very interesting designs to help each one stand out. In the entry-level market, price is king, so designs are minimalist and construction materials are usually cheap. To distinguish the Revo RB-102, Acer has swapped the hard top casing with something a lot closer to carpet.

This fabric-adorned black-shaded box looks fantastic on your desk and does wonders to add a little softness to your office setup, where alternatives would simply serve to harden the overall feel of your workspace. For that reason, you're unlikely to want to hide it by mounting it to the back of your monitor – so bear that in mind. There's nothing distinct about its fairly average size (131 x 131 x 43mm) and weight (449g) otherwise, making it suitably portable if you need to transport the device. Should you need smaller or lighter devices, they certainly exist, but the difference would be negligible.

While we love the overall design, its functionality is a mixed bag. First, let's get the Wi-Fi 6E and Bluetooth 5.3 out of the way – although not the very latest standards, they're an acceptable compromise at this price point.

We were, however, disappointed with the fact that there's only one USB-C port and it's on the front – with no near alternative. This sits alongside a Kensington lock slot, two USB-A 3.2 ports and a 3.5mm audio jack. On the rear, we get two additional USB-A 3.2, DisplayPort 2.1, and HDMI 2.1 ports alongside an RJ45 Ethernet port. There's no room for an SD Card reader, which some devices fit on the side, however. It's certainly usable, but with such a need for USB-C, we'd have loved the flexibility of at least an additional port.

Acer Revo RB-102 review: Specs and performance

The 12-core Intel Core Ultra 5 125U processor is a low-power mobile processor based on the Meteor Lake architecture, with a small capacity for AI workloads thanks to an integrated 11 TOPS NPU. The key appeals here are low power draw, support for multiple displays, and efficiency. This was fitted alongside an Intel Graphics chip, as well as a minimal 8GB RAM and 512GB SSD.

Given this machine's pricing, we wouldn't expect fireworks. Still, performance benchmarking with Geekbench 6 rendered a modest single-core result of 2,093 and a multi-core performance of 6,869. These aren't poor results in isolation – outperforming the comparably priced the Kron Mini K1 (Intel Core i5-12600H) on multi-threaded performance (5,508) but underperforming on competency running the simplest tasks (2,233). Compare it with the excellent Matrix Mini M1, however, with its Intel Core i9-13900HK (14-core) and 32GB RAM producing results of 2,806 and 8,912; this machine is priced at roughly £150 less. Generally, during our experience, the machine ran a little noisy, and navigating across menus and applications wasn't as buttery smooth as we'd have liked.

Similarly, the Iris Xe Graphics chip in that Mini M1 scored 13,514 in our OpenCL test compared with the Acer's 11,029. However, the latter result is respectable and comes in higher than many of the entry-level and even a couple of the mid-range mini PCs we've tested in recent years. Its SSD read and write speeds are also just fine – coming in line at 3,278MB/s and 2,301MB/s respectively. This isn't the quickest but more than passable.

Acer Revo RB-102 review: Is it worth it?

The mid-range market is an incredibly tough nut to crack, with the approaches to both pricing and hardware varying between options.

Instead of opting for a higher-end chip from a few generations ago, as the Acemagic Matrix Mini M1 has done, Acer chose a relatively recent Intel CPU from the first generation of NPU-supported processors to squeeze the smallest amount of AI functionality. So you get features like Windows Studio Effects. But the price is a little higher – and that's despite raw performance coming in relatively cool. That's partially because it's fitted with just 8GB RAM. While it's the latest standard (DDR5 SDRAM vs the DDR4 SDRAM you'll find in most other mini PCs), we'd loved to have seen at least 16GB – but such is the way of the world.

If you can't be swayed by the outstanding Mac Mini, which is supremely more powerful and roughly the same price, the Acer Revo RB-102 is a reasonable Windows mini PC that looks a cut above the rest – and can hold its own against the competition. Chief among its rivals is the lesser Air12 2026 Edition, which the Acer outpaces while offering a bit of AI sauce if you need it. If not, however, you're best looking at the Acemagic Matrix Mini M1, which is cheaper and more powerful.

Acer Revo RB-102 review specifications

AI represents a prime opportunity for security teams to keep pace with accelerating development timelines, according to GitLab CISO Chaim Mazal. But this will require tweaks to traditional processes.

The use of AI in software development is fast approaching near-universal levels, with figures from Stack Overflow’s 2025 Developer Survey showing 84% of developers have adopted the technology.

While AI is helping speed up code production, streamline development lifecycles, and drive productivity, Mazal told ITPro it’s also creating new security risks.

“Businesses have decided that in order to be successful in this next wave of the journey, AI has to be leveraged and teams have to go full bore,” he said.

“I think in tandem and parallel, security teams have been working on figuring out how they can build appropriate solutions to be able to enable their organizations to do that.”

Mazal noted that enterprises are still working around the “same practices and principles that we’ve historically followed” - robust testing, reviews, and security checks.

But with the advent of agentic AI and the pace of change, there’s a risk that blind spots could emerge somewhere in the process.

Indeed, it’s an issue that GitLab recently highlighted in a study. Analysis from the company in June warned that eight-in-ten organizations are adopting AI tools faster than they can develop policies to govern them.

The influx of AI-generated code is also proving troublesome for enterprises, with a study from Tricentis showing that roughly 60% of organizations have shipped untested code.

The result here is that security teams now face a confluence of challenges. Potentially dangerous AI-generated code and increasingly fast development timelines are placing huge pressure on teams to ensure products ship in both a timely and secure manner.

An engineering-first approach

Keeping pace with AI-powered development will mean security teams essentially have to start operating more like their counterparts in development in an ‘engineering-first’ approach.

This is an operating model that takes core engineering principles such as automated testing and CI/CD pipeline techniques and embeds them within security processes. Simply put, rather than acting like manual auditors and policy enforcers, they play a more involved role in development processes.

Moreover, they build parallel products designed to support development from the get-go. Fundamentally, this is about being a “first-level contributor” in software development, Mazal said.

“As our engineering teams move fast, having the security team have the ability to contribute code, make iterative adjustments, and be part and parcel with the development process is key to our success,” he told ITPro.

“Being able to move fast, iterate fast, and not just be consultative in nature, and being able to really roll up our sleeves and work alongside the rest of the teams to build that secure path and the secure guardrails is something I think most modern security teams don’t have the luxury of foregoing.”

When it comes to AI, what’s good for the goose is good for the gander. Security teams should view the technology as a means to keep up with development processes – as with developers, teams need to automate where they can.

Humans in the loop

Ensuring humans remain in the loop throughout the development and testing process will still be vital moving forward, according to Mazal. The exact scope of human involvement is still up for debate, however, and the long-term goal is to automate as much as possible in the safest way

“I think most organizations who make commercial-level software want to get to a place where there are only humans in the loop in places that it’s absolutely necessary, based on the risk to the organization,” he said.

“They’d like to be able to have the overwhelming majority of their code auto-generated and auto-reviewed.”

Mazal said this has the potential to pose new challenges for developers, testers, and security teams alike. Fully-automated pipelines are still evolving, and as ITPro recently reported, some enterprises are exploring the concept of ‘dark testing factories’ to ramp up automation with little human input.

“Collectively we’re still trying to figure out as an industry what those best practices are, but we’re definitely making very big leaps and bounds in very short periods of time.”

GitLab itself is ramping up the use of agents in a software factory approach, Mazal noted, focusing intently on a security-conscious strategy. Agents need strict guardrails, and enterprises need to implement certain restrictions from the get-go.

“How we’re thinking about this at GitLab from a software factory perspective is that you have the ability from the very initial inception of design to bake in those security guardrails and requirements,” he explained.

“You can go ahead and create and craft them and define them based on the data sensitivity and requirements that you have, and you can uniformly implement governance controls across the board.”

These processes will naturally evolve as requirements change, Mazal said, and teams will “continuously make adjustments along the way”.

FOLLOW US ON SOCIAL MEDIA

Poorly implemented IT strategies have a significant impact on businesses. From lackluster growth and financial losses to employee change fatigue, the risks are huge.

With enterprises ramping up AI adoption, many are failing to learn the lessons from previous change programs and are diving headlong into projects.

On this week's episode of the ITPro Podcast, Jane and Ross speak with Al Kingsley MBE, CEO of NetSupport, to explore how IT leaders can implement change, lead by example, and deliver successful transformation projects.

Highlights

“First and foremost, we need to make sure that we're not just making changes because, hey, we've just seen online that everyone's using AI now, so hey, we'll go that way. As opposed to, well, what are we trying to mitigate here? Is it about productivity and output? You know, and the data bit is, well, if we don't know how to robustly interrogate our data, and a bit like those who remember Donald Rumsfeld from the U.S. government, it's not the known knowns or the known unknowns; it's the unknown unknowns. You don't have the right questions to ask, you're not going to get the right answers, and the data is not going to help inform your decision-making.”

“We're seeing more and more agentic AI, AI that's running in parallel and running tasks, being adopted in organizations at pace. But the other thing to balance all that yin and yang with, I think, is the more we're adopting at-pace technology in our organization, it is increasing the value of the human bit. Whether it's the customer service, the engagement with people, the way we think differently, the way we're more creative, and so the real irony is that AI and digital are also amplifying the value of our most human aspects and skills too.”

Links

• Enterprises just can't seem to shake legacy tech – and it’s seriously hampering digital transformation goals
• AI was meant to simplify IT service management – new research shows it's creating bigger workloads for teams

AMD has boosted rack efficiency four-fold since 2024, and the company claims it’s on track to upping it by 20 times by the end of the decade.

The chipmaker revealed it recently exceeded its 2026 target of a three-fold improvement in rack-scale energy efficiency for AI training and inference.

Two 2030 AMD racks are now expected to deliver the same compute as roughly 570 racks did in 2024, enabling a 20x reduction in use-phase electricity and a 28x reduction in carbon intensity.

Alternatively, efficiency gains are expected to enable 20-times more compute, measured in floating point operations per second (FLOPs) per watt, using the same amount of energy.

AMD said energy improved energy efficiency delivers a range of environmental benefits, but also in helping to improve AI performance without increasing power, improving total cost of ownership, and helping customers scale faster.

“The next wave of AI efficiency will depend on tighter co-optimization across compute silicon, memory, interconnects, software and rack-scale system design," said Sam Naffziger, senior vice president and corporate fellow at AMD.

"Our estimated 4x improvement through 2026 reflects the strength of our approach and the progress AMD is making across the full system, putting us ahead of our projected pace toward the 2030 goal.”

The improvements in performance come down to compute capability, memory bandwidth, and interconnect bandwidth, according to AMD.

Advanced process technology and architectural improvements help increase floating-point compute performance per watt, advanced architectures and memory integration improve bandwidth, and high-speed interconnects and tighter integration increase the network bandwidths.

High-bandwidth memory, larger caches, and tighter integration between memory and compute can reduce unnecessary data movement, which uses energy. Meanwhile, high-speed scale-up interconnects help larger systems reduce bottlenecks and share data more efficiently.

AMD’s ROCm software driving gains

Elsewhere, AMD claimed its ROCm software stack and open standards help developers and organizations deploy AI workloads more efficiently on AMD platforms, with open ecosystems also supporting broader optimization across AI and high-performance computing (HPC) deployments.

"The projected rack-scale efficiency gains AMD has set forth can enable AI and HPC customers to achieve data center-level gains in power and cooling infrastructure, which can be limiting factors to increasing compute performance," said the firm.

"Better energy efficiency can help reduce operational electricity use, improve total cost of ownership, and support more sustainable AI and HPC growth."

AMD has also published its 2025-26 Corporate Responsibility Report, detailing its progress across product efficiency, operations, supply chain responsibility, and community impact.

It reduced operational greenhouse gas emissions by 30% from 2020 to 2025, sourced 58% of global AMD electricity from renewable sources and audited 99% of its manufacturing supplier factories for responsible business practices.

"We aim to cut operational greenhouse gas (GHG) emissions in half from 2020 to 2030 as we expand our use of renewable energy. In 2025, 100% of AMD manufacturing suppliers maintained public GHG reduction targets and 82% sourced renewable energy, said AMD chair and chief executive officer Lisa Su.

"The AI era is still in its early stages. Working closely with industry, government and academia, we will keep building the high-performance, energy-efficient compute foundation it requires."

FOLLOW US ON SOCIAL MEDIA

AI cloud firm Nebius is leasing high-density capacity at Vantage Data Centers' Newport campus, marking the first commercial capacity commitment in the South Wales AI Growth Zone.

Under the agreement, Nebius will lease Nvidia-powered capacity at Nebius's CWL1 facility to support AI training, inference, agentic AI and enterprise AI workloads, and serving enterprises, researchers, start-ups and public sector organizations.

“This is an important milestone for South Wales and for the UK’s AI infrastructure ambitions,” said David Howson, president, EMEA, Vantage Data Centers.

“Nebius is scaling AI cloud infrastructure in the UK at a time when demand for domestic compute capacity continues to accelerate. Vantage’s Newport campus provides the scale, connectivity and operational excellence needed to support that growth, and we’re proud to help establish South Wales as a leading destination for AI investment.”

One of Europe’s largest data center campuses, Nebius's CWL1 site opened in 2010. It runs on 100% certified renewable energy, with its latest facilities designed to minimize water consumption through a closed-loop cooling system that recirculates water rather than evaporating it.

The facility also uses air-cooled chillers and free cooling when ambient conditions permit.

The increased capacity forms part of Nebius’ broader UK expansion and commitment to scale domestic AI compute capacity.

In June, the company announced around £1.7 billion of committed capacity build-out across four UK sites, including the delivery of more than 1GW of AI-ready capacity across Newport, Bridgend, and Bro Tathan.

“The UK is one of the places where AI is being built, deployed and adopted by startups, enterprises and the public sector,” said Gary Tierney, general manager for EMEA at Nebius.

“Vantage’s Newport campus gives us a strong foundation to expand access to NVIDIA-powered AI infrastructure in the UK and support customers building the next generation of AI applications.”

AI Growth Zones take shape

The buildout is based on Nvidia's latest full-stack, end-to-end DSX AI factory reference design, and supports the UK government’s AI Opportunities Action Plan.

South Wales was designated as one of the UK's UK AI Growth Zones in November last year. At the time, the government said it could create more than 5,000 new jobs for local communities.

It's also home to the UK’s largest cluster of semiconductor businesses.

The AI Growth Zone stretches along the M4 corridor from Newport to Bridgend has the potential to harness over 1GW by early 2030s. 

“Today’s announcement marks an important milestone for the South Wales AI Growth Zone and demonstrates Wales’ growing reputation as a strategic location for the infrastructure that will power the AI economy,” said Adam Price, Welsh cabinet minister for enterprise, connectivity and energy.

"Nebius and Vantage’s investment will help unlock new economic opportunities, support highly skilled jobs and reinforce Wales’ position as a leading destination for advanced digital investment."

The move by Nebius comes hot on the heels of AI Growth Zone announcements elsewhere across the UK this week. The Lanarkshire growth zone in Scotland, for example, secured £300 million in investment to ramp up data center development, with Dell Technologies pledging its support for the site.

FOLLOW US ON SOCIAL MEDIA