OpenAI Podcast

Jason Wolf, an OpenAI researcher on alignment who shapes model behavior and the Model Spec. He explains what the Model Spec is and how it guides model decisions. Topics include the spec’s structure and updates, the hierarchy for conflicting instructions, handling edge cases like Santa Claus, the role of chain-of-thought, and how the spec evolves with new capabilities and feedback.

Karan Singhal, who leads Health AI Research and focuses on model evaluation and safety; Dr. Nate Gross, a physician leading Health efforts with experience in policy and clinical care. They discuss training models for sensitive health questions, HealthBench and physician-guided evaluation, tailoring responses to patients and clinicians, integrating records and devices, and real-world deployments and measurement challenges.

How should advertising work in an AI product? Asad Awan, one of the ad leads at OpenAI, walks through how the company is approaching this decision and why it’s testing ads in ChatGPT at all. He explains how ads are built to stay separate from the model response, keep conversations with ChatGPT private from advertisers, and give people control over their experience.Chapters00:00:29 — Mission and principles00:04:01 — Separation between ads and answers00:07:31 — Who will see ads00:08:52 — Internal input and decision-making process00:11:06 — Controls and how ads will work00:15:53 — Guardrails for sensitive conversations00:17:33 — Skepticism about ads00:20:26 — Helping small businesses00:24:13 — Future of ads Hosted on Acast. See acast.com/privacy for more information.

OpenAI CFO Sarah Friar and Khosla Ventures founder Vinod Khosla argue the greatest challenges in AI right now are keeping up with demand and making sure more people get the benefit. They unpack what's driving big investments in compute and why this moment is different from other technology cycles — with meaningful advances in health, agents, and robotics still ahead. Chapters00:00:00 — What’s the AI story of 2026?00:07:28 — AI in healthcare00:12:01 — Scaling compute to match revenue00:18:05 — Difference between now and dot-com bubble00:27:41 — Ads in ChatGPT00:30:05 — Will consumers have more than one AI subscription?00:36:41 — Winning in enterprise00:39:44 — How can startups succeed?00:44:05 — Robotics and beyond Hosted on Acast. See acast.com/privacy for more information.

Christina Kim, a research lead at OpenAI focusing on model reasoning and emotional intelligence, joins product manager Laurentia Romaniuk, who oversees model behavior and user experience. They discuss the concept of AI 'personality' and the advancements in GPT-5.1, particularly its ability to reason and follow user instructions. The duo dives into how user feedback shapes the model, measuring emotional intelligence, and balancing personality with steerability. They also explore future possibilities for customization and improving user interactions.

Kevin Weil, Head of OpenAI for Science, and Alex Lupsasca, an OpenAI research scientist and physicist, explore how GPT-5 accelerates scientific discovery. They discuss its role in literature searches and solving complex physics problems like black hole symmetries. The duo emphasizes GPT-5’s capacity to connect diverse research topics and its application in fusion experiments. They predict significant advancements in science over the next five years, while advising researchers to embrace AI as a collaborator for innovative explorations.

How will the internet feel when your browser can actually help do things for you? OpenAI’s Ben Goodger and Darin Fisher, whose past work shaped some of the most popular modern browsers, dive into the making of ChatGPT Atlas. They explore how AI changes what a browser can be, from tabs you can talk to, to agents that take over tedious tasks. Learn more about the decisions they made along the way and what’s coming next.- 00:00:45 What is Atlas?- 00:03:34 The state of browsers and AI on the web- 00:13:55 Under the hood: why browsers are hard (OWL, rendering)- 00:22:00 Building with AI: Codex, cross-language, Swift on Windows- 00:33:39 Search in Atlas: one box plus model response- 00:41:28 Favorite features: scrolling tabs and tab search- 00:45:23 Side Chat in action: summarize, shop, build forms- 00:46:59 Real-world wins with Agent (cloud bill, medical results)- 00:52:45 Why Chromium? Compatibility and extensions- 01:07:57 Five-year vision: an agentic web and reduced toil- 01:13:11 Power tips and closing remarksLearn more about OWLhttps://openai.com/index/building-chatgpt-atlas/ Hosted on Acast. See acast.com/privacy for more information.

Sam Altman and Greg Brockman from OpenAI sit down with Broadcom’s Hock Tan and Charlie Kawwas to discuss their new partnership—and what it means for the future of AI. From custom silicon to global-scale infrastructure, they share how compute innovation is shaping the road to AGI. 00:00 Announcing the partnership03:06 The scale of AI infrastructure06:03 Collaboration and innovation in chip design08:49 Historical context and future vision12:10 Role of compute in AI development15:01 Optimizing for specific workloads18:02 Journey towards AGI21:00 Future of AI and compute capacity23:50 Wrap-up and future projects Hosted on Acast. See acast.com/privacy for more information.

The OpenAI Podcast is live for the first time. Host Andrew Mayne sits down with startups Cursor, Abridge, SchoolAI, and Jam.dev—each reimagining how AI can transform their industries. From healthcare and education to coding and collaboration, we explore how these builders are putting AI to work in the real world.00:23 Caleb Hicks (SchoolAI)14:14 Dani Grant (Jam.dev)26:20 Zach Lipton (Abridge)44:38 Lee Robinson (Cursor) Hosted on Acast. See acast.com/privacy for more information.

Greg Brockman, co-founder and president of OpenAI, and Thibault Sottiaux, Codex engineering lead at OpenAI, discuss the evolution of coding with AI. They dive into the journey from GPT-3 to the powerful GPT-5 Codex, exploring its capabilities in complex refactoring and coding assistance. The duo reveals insights on building AI harnesses, breakthroughs in code review, and the transformative potential of agentic coding. They also touch on the future landscape of AI in software development and the implications for learning to code in an automated world.