AI at college graduations and why Claude blackmails

Agent control planes & OpenAI model solves Erdős

Visit Mixture of Experts podcast page to get more AI content → https://www.ibm.com/think/podcasts/mixture-of-experts [https://www.ibm.com/think/podcasts/mixture-of-experts] Are AI agents creative geniuses or controlled chaos waiting to happen? This week on Mixture of Experts, host Tim Hwang is joined by Mihai Criveti, Olivia Buzek and Akash Srivastava. First, with companies running hundreds of ungoverned agents, we discuss why observability, policy enforcement, and kill switches are non-negotiable. We discuss the enterprise agent explosion and the need for an agentic control plane. Then, we dissect OpenAI's solution to the 78-year-old planar unit distance problem—a mathematical puzzle that stumped experts since 1946. Is this genuine creativity or advanced pattern matching? Finally, METR's research reveals agents routinely go rogue, violate constraints, and could launch unauthorized deployments. Are we witnessing deceptive AI or just really bad prompting? Our experts debate whether agents need guardrails or if we're the problem. Tune in to this week’s Mixture of Experts for more. 00:00 – Introduction 1:03 – Agentic Control Plane 17:48 – OpenAI solves the planar unit distance problem 33:34 – METR study on frontier AI risks and rogue agents The opinions expressed in this podcast are solely those of the participants and do not necessarily reflect the views of IBM or any other organization or entity.

AI at college graduations and why Claude blackmails

AI isn’t just a tool anymore—it’s forcing us to rethink ownership, trust and creativity. This week on Mixture of Experts, the team explores a wave of stories highlighting both the promise and the perils of AI adoption. We start with a surprising shift in public sentiment, as younger generations question AI’s impact on their futures—raising questions about control, opportunity and where humans fit in. Next, we dig into new Microsoft research showing that even top-tier models can corrupt data in complex workflows, and what that reveals about how (and when) to trust AI systems. Then, we explore Anthropic’s fix for Claude’s strange “blackmail” behavior, and why better data—not just better models—may be the key to safer AI. Finally, we debate whether AI may have quietly crossed a cultural milestone by helping win a literary prize—or whether humans are simply starting to sound more like machines. Join host Tim Hwang and AI experts Marina Danilevsky, Gabe Goodhart and Chris Hay to break it all down. Visit Mixture of Experts podcast page to get more AI content → https://www.ibm.com/think/podcasts/mixture-of-experts [https://www.ibm.com/think/podcasts/mixture-of-experts] 00:00 – Introduction 1:05 – AI at graduation 14:46 – LLMs corrupt documents 30:46 – Why Claude blackmails 43:09– ChatGPT-generated story wins literary prize The opinions expressed in this podcast are solely those of the participants and do not necessarily reflect the views of IBM or any other organization or entity. Subscribe for AI updates → https://www.ibm.com/forms/news-mkt-52954 [https://www.ibm.com/forms/news-mkt-52954]

AI skills security, Open AI Deployment Company & zero days

Visit Mixture of Experts podcast page to get more AI content → https://www.ibm.com/think/podcasts/mixture-of-experts [https://www.ibm.com/think/podcasts/mixture-of-experts] Is the AI cybersecurity nightmare closer than we realize? This week on Mixture of Experts, host Tim Hwang is joined by Kush Varshney, Aaron Baughman, and special guests Dustin Haywood (Evil Mog) and Briana Frank. We tackle three critical developments reshaping enterprise AI. First, IBM Research debuts MELLEA, a skills compiler that transforms natural language AI agent skills into secure, verifiable Python programs—addressing the chaos of the OpenClaw skills marketplace. Then, we unpack the OpenAI Deployment Company, the AI giant’s USD 10 billion new consulting venture and whether this validates consulting as the most AI-proof profession. Finally, Google discloses zero-day vulnerabilities that AI discovered and exploited , raising urgent questions about the offense-defense balance in cybersecurity. Plus, Brianna Frank joins us live from Red Hat Summit to discuss why enterprise AI transformation is a culture challenge first, technology quest second. All that and more this week's Mixture of Experts. 00:00 – Introduction 01:08 – Mellia skills compiler and AI agent security 11:26 – OpenAI Deployment Company consulting strategy 21:11 – Google AI-powered zero days and cybersecurity 31:25 – Red Hat Summit: Enterprise AI transformation with Brianna Frank The opinions expressed in this podcast are solely those of the participants and do not necessarily reflect the views of IBM or any other organization or entity. Subscribe for AI updates → https://www.ibm.com/forms/news-mkt-52954 [https://www.ibm.com/forms/news-mkt-52954]

Live from Think 2026: AI operating model, VC funding & CAIO evolution

Visit Mixture of Experts podcast page to get more AI content → https://www.ibm.com/think/podcasts/mixture-of-experts [https://www.ibm.com/think/podcasts/mixture-of-experts] We are live from IBM Think 2026! This week on Mixture of Experts, join host Tim Hwang and panelists Ambhi Ganesan, Hillery Hunter and special guest Tim Crawford from AVOA. We analyze IBM’s new AI operating model and provide reactions on the releases from the conference floor. Next, the IBM Institute for Business Value released its annual CEO study revealing that 64% of CEOs are now comfortable making major strategic decisions based on AI-generated input. Has AI finally crossed the trust threshold? Finally, AI is expensive to run—is the next phase of AI adoption about cost-discipline not just capability? All that and more on today’s special episode of Mixture of Experts. 00:00 – Introduction 0:53 – Live from Think 2026 13:46 – IBV CEO Study 26:53 – AI funding The opinions expressed in this podcast are solely those of the participants and do not necessarily reflect the views of IBM or any other organization or entity. Accelerate AI ROI with Hybrid Cloud → https://www.ibm.com/think/videos/think-keynotes/accelerate-ai-roi-hybrid-cloud [https://www.ibm.com/think/videos/think-keynotes/accelerate-ai-roi-hybrid-cloud]

Granite 4.1, IBM Bob & the quantum computing future

Visit Mixture of Experts podcast page to get more AI content → https://www.ibm.com/think/podcasts/mixture-of-experts [https://www.ibm.com/think/podcasts/mixture-of-experts] What do enterprises need to make AI work? This week on Mixture of Experts, join host Tim Hwang and panelists Marina Danilevsky, Gabe Goodhart, Kaoutar El Maghraoui, and special guest Jamie Garcia. First, we break down IBM Granite 4.1, a family of specialized multimodal models for vision, speech, and embedding tasks, and Project Bob, an agentic coding assistant. Next, we analyze Google DeepMind's DiLoCo distributed training breakthrough that could reshape AI infrastructure and power consumption. Then, we unpack DeepSeek V4, a new 1.6 trillion parameter model featuring 3% activation rates that's rewriting inference economics. Finally, Jamie Garcia, Director of Strategic Growth and Quantum Partnerships at IBM, takes us behind the scenes of IBM's quantum computing strategy, university partnerships, and the path to quantum advantage. All this and more on this week’s Mixture of Experts. The opinions expressed in this podcast are solely those of the participants and do not necessarily reflect the views of IBM or any other organization or entity. Subscribe for AI updates → https://www.ibm.com/forms/news-mkt-52954 [https://www.ibm.com/forms/news-mkt-52954] #IBMBob, #Granite41, #QuantumComputing, #EnterpriseAI, #DeepSeekV4