Peter Orbán

Human or AI?: The Nuances of Intelligence The differences between humans and AI are not just within the breadth of abilities that humans demonstrate but also in unique and critical skills that AI programs have yet to exhibit. As we continue to unpack what AI means for work, we should not overlook the intrinsic value of these qualities, which are critical to a healthy world and workplace—and uniquely human. #Human #Leadership #AI #KornFerry

25

Amazon Web Services (AWS) for Media and Entertainment: Leaders Discussion: Levers for Success in 2025. Chair Ian Munford, Industry Development Lead MEGS, EMEA, AWS Samira Panah Bakhtiar, GM MEGS Dual AI economy emerging. Those leveraging and those only doing POCs. Anthropic could work on task for 7 hours without losing context with Model Context Protocol. Agents become strategic. Agents to understand rights of content. Where can it be sold to new distribution partners. Increase reach of content. AI pace advances quickly. Ian Whittaker, Analyst, Liberty Sky Advisors What is a view? People monetising while debate continues. Definition and standards vary with broadcasters and tech.

23

2 Comments

Artificial Intelligence: 10 times faster and more impactful than the Industrial Revolution? That’s the vision shared by Demis Hassabis CEO of Google DeepMind and The Nobel Prize laureate, who foresees the arrival of Artificial General Intelligence (AGI) within the next 5 to 10 years. Such a shift could transform healthcare, energy, scientific discovery and more. For Africa, this brings both opportunities and challenges: • Opportunities to improve healthcare access, strengthen education, and foster local innovation. • Challenges to ensure benefits are fairly shared, to avoid becoming just a testing ground, and to anticipate the impact on jobs and inequalities. As Hassabis describes himself, being a “cautious optimist” is key: embracing the potential of AI while building the right safeguards and governance. 👉 The future of AI will not be defined only in the labs of global tech giants it also depends on how regions like Africa shape it into a tool for inclusive and sustainable development.

16

7 Comments

My guests in the latest episode of the Superposition Guy's Podcast are the co-founders of Planckian Michele Dallari, CEO, and Marco Polini. Spotify: https://lnkd.in/di3xRE7g Trancript: https://lnkd.in/da3AF7nn We talk about their company’s novel superconducting chip architecture, which uses global control and always-on ZZ interactions to drastically reduce wiring complexity and improve scalability. Marco and Michele discuss the advantages of controlling many qubits with just two or three classical lines, their early-stage experimental progress, and the software adaptations required to compile algorithms for their architecture. The conversation also touches on Italy’s growing quantum ecosystem, Planckian’s collaborative approach to chip manufacturing, their excitement about recent advances in pulse engineering to compensate for fabrication imperfections—and much more.

19

1 Comment

Meta "Explicitly Separating" #Horizon Worlds from #Quest's Store! Meta published a blog post for developers wherein it lays out its new strategy for the Quest ecosystem and Horizon Worlds, taking the two in separate directions. Concretely, this means that Meta is going to focus much more on 3P instead of 1P, and that could very well turn out to be a very strong strategy!

17

1 Comment

Adobe has launched Adobe AI Foundry (https://lnkd.in/efaQ_z_3), a service that lets enterprise brands build custom generative AI models trained on their own visual, audio and video assets. Adobe is by no means the first to launch such a service, which it has built by enabling Foundry to fine-tune existing Adobe Firefly models for enterprises. Microsoft already offers Azure AI Foundry, although its main focus is broader AI applications and agent development rather than brand-asset-specific generative model training. What it signals is yet another evolution in how marketing teams can leverage AI, moving teams from ‘users’ of external AI tools, with all the built-in risks and questions around IP and creative ownership, to ‘builders’ of the AI engine that can sit behind a brand. It also shifts the skillsets required to leverage AI, from marketers who know how to use tools, to marketers who know how to build and steward brand-specific models. Skills like model training, governance and workflow integration, not just ‘prompt engineering’ will be needed. So how can you get started? Well, Adobe AI Foundry is an enterprise offering at the moment, so for most brands it’s not an option. But the starting point for developing your own generative AI capability is the same whatever platform you use. First, you need something to train a model on, so you need to audit your brand’s asset and media library and ask questions such as how usable is it as training data for an AI model? Is it mainly structured or unstructured data for example. Once you’ve done this, you’ll know what and how much data you have to work with. Second you need to consider how to embed this new capability into your marketing team, re-thinking your existing workflows. E.g. who curates the prompt, who reviews outputs, how does governance work, where is the human in the loop? Then you need to test (suggest social visuals or maybe short form video ads) and measure at all stages. What is the time taken, what is the cost, what is quality, what is the consistency? Only then will you be able to see the value when compared to your existing approach. #MarTech #GenAI #CreativeAutomation #BrandStrategy #MarketingInnovation

6

2 Comments

Live from CES: 1. Nvidia launched its Vera Rubin platform across six new chips, saying Vera Rubin will offer dramatic reductions in inference and training costs compared to Blackwell: https://lnkd.in/gt34H--k 2. Nvidia rolled out DLSS 4.5 with new 6x Multi Frame Generation for its RTX 50 series, and a second-generation Super Resolution transformer model for all RTX GPUs: https://lnkd.in/gH_rQBgx 3. Nvidia announced native apps for its GeForce Now cloud gaming service for select Linux distributions and Amazon Fire TV sticks, available in early 2026: https://lnkd.in/gTMhYpfm 4. Nvidia released its Alpamayo family of AI models, tools, and datasets for autonomous vehicles, and detailed a collaboration with Mercedes-Benz on its first full-stack autonomous vehicle effort: https://lnkd.in/gD3tagGw 5. Nvidia said that it's seeing “strong” demand from China for its H200 chips but is awaiting approvals from both Washington and Beijing before sales can begin: https://lnkd.in/gxsZTrgw 6. Nvidia partnered with UMG “to pioneer responsible AI for music discovery, creation, and engagement”, centered on Nvidia's Music Flamingo model: https://lnkd.in/gg6uEH9W

16

3 Comments

#CES2026: From #DigitalAI to #PhysicalAI CES 2026 felt fundamentally different from CES 2025. Last year’s focus was largely on AI’s potential—generative models, copilots, and productivity tools. The energy was exploratory and software-centric. CES 2026 had a different focus: Where is AI already acting in the physical world and at scale? NVIDIA's Jensen Huang’s keynote moved beyond model intelligence to AI systems that can perceive, reason, and act. Agentic AI, robotics, simulation, and accelerated computing were presented as integrated, operational systems, not future concepts. Compared to 2025, it was clear that intelligence alone is insufficient; value is created when AI can execute in the real physical world. One example of this manifestations was Jensen's keynote with Siemens CEO Roland Busch, which highlighted industrial physical AI at scale: connecting digital twins, automation, and AI-driven decision-making to factories, infrastructure, and supply chains. It reinforced that some of the most immediate returns from physical AI will come from industrial systems, not just consumer applications. Other examples include autonomous vehicles and humanoid robots, which also felt more grounded this year. The focus shifted from spectacle to deployment readiness—safety, reliability, and unit economics—particularly in manufacturing, logistics, and service environments. One implication was the accelerating displacement of physical labor. Unlike prior automation that focused on repetitive tasks, physical AI is expanding automation into semi-structured and previously human-only roles. Over the next decade, this will reshape labor markets in ways that are both economically powerful and socially consequential. Key takeaways: • Physical AI will change cost structures not just productivity Robotics and autonomy are moving from CapEx-heavy experiments to scalable operating models, fundamentally altering labor economics. • Talent strategy must evolve alongside automation strategy Winning companies are investing early in reskilling: transitioning workers from manual roles into supervision, maintenance, system orchestration, and human–AI collaboration. • Operational resilience becomes a workforce issue As physical AI systems scale, organizations will depend on fewer but more highly skilled operators. This increases the importance of training, redundancy, and trust in AI-driven operations. • Displacement will be uneven Physical labor displacement will disproportionately affect frontline and hourly workers. The question is not job loss vs. job creation, but timing. New roles will emerge, but often not fast enough, or in the same locations, as jobs are displaced. This gap is where tension will surface Physical AI will unlock extraordinary possibilities, but it will also test how responsibly organizations lead through disruption and how thoughtfully they scale tech alongside people. PS. It's also great to see so many friends, thank you!

70

5 Comments