Nitin Aggarwal

Living amid constant uncertainty, we must plan for different scenarios for the short, medium, and long term. Hansa Iyengar discusses how enterprise leaders must architect their technology portfolios and operating models for optionality, speed, and resilience HFS Research Saurabh Gupta Tony Filippone Dana Daher Don Ryan

63

3 Comments

Had the absolute pleasure of sitting down with Sriram Raghavan, VP of AI Research at IBM, for the latest episode of Ctrl + Alt + AI, and I learned so much. Sriram has spent over two decades at IBM shaping the direction of AI research. In this episode, we went deep into how enterprise AI is evolving, and what it really takes to build responsible, efficient models that can scale in the real world. A few things we covered: ✅ The surprising performance of IBM’s Granite models, despite being smaller than most foundation models ✅ Why IBM chose to open-source them, and what they've learned from engaging with the OSS community ✅ What makes enterprise data readiness so hard, and how it changes the game compared to internet-scale models ✅ How Guardian AI helps ensure AI systems stay safe and aligned ✅ And Sriram’s take on what comes after general-purpose LLMs, including his excitement around neuro-symbolic reasoning and causal models We also talked about the gap (and opportunity!) between academia and industry, and what both sides can learn from each other in this next chapter of AI. If you’re building or investing in enterprise AI, or just want a peek behind the curtain of one of the most ambitious AI research orgs, this is a conversation worth watching. 🎥 Full episode is live now on my YouTube channel. Would love to hear your thoughts!

143

21 Comments

Shweta Shah, Ph.D. has suggested a concept that she calls "Content Quality Management". Shweta and I have been writing together for a while, and I am really impressed by this idea, that I think addresses exactly what the data community needs to consider, going forward in an era so indisputably dominated by AI. Shweta is adding to an ecosystem of thinkers like Bill Inmon (with his absolutely wonderful concept of the "textual warehouse") and Jessica Talisman (coiner of the "ontology pipeline"), and a wealth of people in the records- and information management space like tech entrepreneur Rachael Greaves, and skilled practitioners like Michael Upton, Andy Potter (deeply involved in ISO standards on AI records), and Stephen Clarke MRIM that are trained in thinking in the intersection of language and data. If you want ideas in how to succeed with AI, I suggest you follow Shweta Shah, Ph.D. #contentquality #contentqualitymanagement #ai #data #language #dataquality #dataqualitymanagement

42

6 Comments

While India builds its large foundation models, there is a lot of value to be had also from small models if we use the right tools and techniques to get the best out of them. At IBM Research India we have been working on this premise and bringing our innovations into the open for the ecosystem to benefits. Here is a great example of work from my colleagues at IBM Research India - applying some of IBM’s open source tools (SDG_Hub, fms-hf-tuning) and some of our research to IBM Granite and BharatGen’s Soverign Indian LLMs to show how this can work and provide significant uplift in performance. https://lnkd.in/gqgjDuJb This was also a tool of discussion in the recent Carnegie India GTS Innovation Dialogue where I spoke about the need for the Indian AI community to now develop tools and solution architectures that enable us to drive value at scale from the investment in Infrastructure and model training. Let’s build Artificial Useful Intelligence (AUI) for India. IBM’s solution kit for BharatGen models will bring a lot of this scaling tech to India’s AI solution builders, out of the box. #ibmresearchindia #indiaai #aui Sandip Patel Ajay Mittal Rishi Bal Ganesh Ramakrishnan Sriram Raghavan Shalini Kapoor Praveen Jayachandran Sachindra Joshi

101

2 Comments

What brought together Nvidia, AMD, Snowflake, Databricks and other financial investors to invest $260 million in Uniphore ? Uniphore co-founder and CEO Umesh Sachdev puts himself in their shoes and says that the strategic investors perhaps see the start-up as a key partner to drive AI adoption “Whether you’re a compute provider or data provider, ultimately, the usage of your technology happens when agentic use cases go live. They see Uniphore as a very important partner to drive business AI adoption,” he said. With ARR crossing $200 million and the way the TCV (Total Contract Value) is growing, Sachdev says going for an IPO “naturally presents itself as an option” in the “not-too-distant future” For the entire conversation, check out the link in comments

28

2 Comments

🚨 A new Data Radicals episode is live — and this one’s a masterclass in data products for AI. I sat down with Sanjeev Mohan — former Gartner analyst, author of Data Products for Dummies, and host of the It Depends podcast — to unpack what data products really are, how they fuel trustworthy AI, and why AI agents will reshape the future of work. Here are 3 insights that hit home: 🔁 Data products provide a "contract" between data producers and consumers. This builds trust and reuse. With ownership, documentation, and discoverability baked in, data products break down silos and finally make data usable – for data consumers and AI models alike. 💡 Data products bring software thinking to data. “We always buy software with a version number… but when was the last time we ever talked about data having a version number?” Data products shift the focus from raw artifacts to trusted, managed assets. 🤖 Your AI projects may be failing because they lack a data-product foundation. Running AI agents atop trusted data products boosts reliability and mitigates hallucination risk. As Sanjeev points out, “Data products give us the trust layer. What if I put my agents, my assistant, my chatbot on top of a data product?” Want to innovate with AI? You’ve got to experiment. As Sanjeev put it: “Unless you experiment, how do you know?” 🎧 Tune in and learn why Sanjeev calls this "the golden age of data": 👉 https://lnkd.in/gw7655GB #DataRadicals #AI #DataProducts #AIagents #DataGovernance #Metadata #FutureOfWork #Podcast #LLMs #DataLeadership

64

4 Comments

I appreciated spending Monday afternoon with IBM CEO Arvind Krishna and most of his leadership team discussing… enterprise AI. I can’t give verbatims as they were NDA conversations, but here’s the upshot from Arvind: He emphasized AI technology's growing significance in an enterprise’s competitive advantage, noting tech's role in globalization, supply chains, and cybersecurity. IBM wants to help its customers deploy AI, focusing on three areas: client AI deployment, internal AI governance, and domain-specific AI models. Arvind highlighted IBM's partnerships with OpenAI and Anthropic and its investments in quantum computing, aiming for practical applications by the end of the decade. He also discussed the need for IBM to improve its narrative on open source and the importance of fostering a risk-taking culture to drive innovation.   My take: I think IBM will be one of the large enterprise AI winners, particularly in industries that are highly regulated. Its biggest, long-term bet is in quantum where it could be THE leader, not just A leader, given its resources to scale, focused end customers, and history. Many think they it will upset today’s GPU-XPU market. I’m not there yet but keep your eye on 2030. Overall, I think the company should rethink messaging or execution of the message it delivers, or will deliver, more than people realize. Arvind acknowledged the “open” messaging. Did you know IBM is the *only* company that offers a hybrid, multicloud AI platform spanning on-prem legacy and private-sovereign-public cloud? This is what large enterprises want a few are discussing this in my circles. The MIS team is at IBM TechXChange this week getting a download on the important announcements. Follow them for updates.

334

11 Comments

New GenAI foundation models are released every two and a half days, yet nearly half of CIOs report that AI hasn't met ROI expectations: https://gtnr.it/4eb0PPL Navigate this dichotomy by determining your organization’s AI pace — and taking the right next steps ⬆️ Gartner for IT | #GartnerIT #GenAI #CIO

13

✨ TechSOCIAL | Exploring the AI Era with Soma Dhavala ✨ On Friday, we had the privilege of hosting Soma S Dhavala for a powerful session on the evolution of AI and large language models (LLMs)—from Software 1.0 to the current 3.0 era, where prompts and context are driving intelligent systems. Soma unpacked the realities of designing with AI, emphasized the importance of human oversight, and challenged us to think of AI as an institution—one that must be governed with responsibility. We explored: 🔹 AI’s evolution across 3 software paradigms 🔹 LLMs as operating systems powering tools, memory, and reasoning 🔹 Why context engineering beats one-off prompts 🔹 The mantra: "Let AI act, let humans decide" 🔹 The hidden costs and trade-offs of AI at scale A huge thank you to Soma for his clarity and generosity, and to everyone who joined the session and contributed to a thought-provoking discussion. Check out the key learnings and session snapshots below! #TechSOCIAL #AIForSocialImpact #ResponsibleAI #LLMs #DigitalTransformation #Software3.0 #ContextEngineering #AIProductivity #TechForGood Digital Bharat Collaborative Piramal Foundation

70

4 Comments

It was an absolute treat to moderate a panel with veteran pricing experts Madhavan Ramanujam and Joshua Bloom who had a wealth of insights and “hot takes” to share on Pricing for Ai products. This is a topic that’s become top of mind for early and mature companies alike and has certainly been a part of my board level discussions. Here were some of my more interesting takeaways from the conversation: 1. Never before has understanding value been more important. Early “willingness to pay” conversations are even more of a MUST during product development. 2. For determining a product's ideal pricing model, think of a 2x2 matrix based on levels of autonomy and attribution: • Low Autonomy + Low Attribution = Seat-Based/Subscription (eg. Saas)   • Low Autonomy + High Attribution = Hybrid (eg. Ai coding)  • Autonomous + Low Attribution = Usage-Based (eg. Infrastructure)  • Autonomous + High Attribution = Outcome-Based (eg. Customer support) 3. Finance teams and Investors should stop focusing on ARR (Annual Recurring Revenue) and start focusing on ARRR (Annual Recurring Run Rate). There is inherently less predictability in value based pricing for Ai and companies will be continuously optimizing through price discovery and commercial negotiations. Interested in learning more? Check out Madhavan’s upcoming book on Amazon: Scaling Innovation.

55

1 Comment