The “godfather of AI” on the risks that worry him most | Geoffrey Hinton
Geoffrey Hinton helped create the neural networks behind today’s AI, won a Nobel Prize for it — and has spent the years since warning about where it’s headed. In conversation with journalist Nicholas Thompson, Hinton shares the AI risks he’s worried about, sketches out a best-case future for the technology and explores whether or not chatbots have subjective experience. Plus: hear his thoughts on the one rule we must give superintelligent AI. Hosted on Acast. See acast.com/privacy for more information.
Why PMs are suddenly valuable again | Tulsee Doshi (Head of Product, Gemini)
Tulsee Doshi is head of product for Google’s frontier AI models, including Gemini and other multimodal models, such as Nano Banana, Genie, Veo, and Lyria.
We discuss:
1. Why model development is not just engineering, but also like Snape’s potions class
2. Google’s full-stack frontier AI strategy
3. What PMs do inside a frontier AI lab
4. Tulsee’s 60-20-20 framework for balancing what works now and what’s almost working
5. How to write your first eval
6. The rising value of great PMs
—
Brought to you by:
WorkOS—Make your app enterprise-ready, with SSO, SCIM, RBAC, and more
DX—Engineering intelligence for the AI era
—
Episode transcript: https://www.lennysnewsletter.com/p/what-it-actually-takes-to-build-frontier
—
Archive of all Lenny's Podcast transcripts: https://www.dropbox.com/scl/fo/yxi4s2w998p1gvtpu4193/AMdNPR8AOw0lMklwtnC0TrQ?rlkey=j06x0nipoti519e0xgm23zsn9&st=ahz0fj11&dl=0
—
Highlights: https://lennyspodcast.com/mostreplayedmoments
—
Where to find Tulsee Doshi:
• X: https://x.com/tulseedoshi
• LinkedIn: linkedin.com/in/tulsee-doshi
—
Where to find Lenny:
• Newsletter: https://www.lennysnewsletter.com
• X: https://twitter.com/lennysan
• LinkedIn: https://www.linkedin.com/in/lennyrachitsky/
—
In this episode, we cover:
(00:00) Teaser
(01:08) What people misunderstand about how models are built
(06:28) Google’s strategy in the AI race
(09:57) AI safety, monitoring, and strengthening cyberdefenses
(14:10) Balancing fast launches with great user experiences
(17:02) Why the model itself has a UX
(19:21) Why product managers are becoming more valuable
(21:24) The skills that matter for research PMs
(23:04) How to build your first eval
(24:37) Getting better at judging a model’s vibes
(26:15) Improving skin tone representation with better evals
(28:25) Evaluating the full experience, not just the model
(30:38) From crowdsourced flash teams to agent swarms
(33:13) The 60-20-20 rule for building ahead of model capabilities
(35:46) Keeping your product adaptable as models improve
(36:42) What an AI harness actually is
(37:30) Upskilling PMs through shared skills and workshops
(41:18) Building the habit of asking, “Can AI do this for me?”
(42:38) Google’s opportunity to build a personal AI assistant
(44:10) Burnout, meaningful work, and the pressure to do more
(47:44) Avoiding AI slop
(49:30) How AI is changing the PM role
(51:02) Lightning round and final thoughts
—
References: https://www.lennysnewsletter.com/p/what-it-actually-takes-to-build-frontier
—
Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email podcast@lennyrachitsky.com.
—
Lenny may be an investor in the companies discussed.
To hear more, visit www.lennysnewsletter.com
Open to Debate: should we trust AI agents with our information?
Would you hand a stranger your bank passwords, your private messages, and control of your personal information? That’s the ask when the stranger is an AI agent, like Meta Muse, Claude Duet, or OpenAI Dots. In this special episode from Open to Debate, recorded live at the Masters of Scale Summit in San Francisco on October 22, 2026, Jeff Berman, co-host of Masters of Scale, moderates a debate on the question: should we stop trusting AI agents with our information? Arguing yes, Meredith Whittaker, president of Signal, says the convenience comes at the cost of our privacy, since agents need sweeping access that could undermine encryption, expose personal data, and create new vulnerabilities for hackers. Arguing no, Neil Chilson, head of AI policy at the Abundance Institute, counters that existing laws, competition, and targeted safeguards can manage the risks without giving up the productivity gains. So how much control is too much?
Follow Open to Debate on YouTube (https://www.youtube.com/@OpentoDebate)
This episode originally ran Friday, November 6th on the podcast “Open to Debate.”
See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.
Eric Schmidt scores US vs. China: who’s winning the AI race?
The AI race between the US and China is the defining competition of our age. Eric Schmidt, former CEO of Google, advisor to the Pentagon, and current CEO of Relativity Space joins Rapid Response live from Masters of Scale Summit to give his most unvarnished assessment yet. He reveals that the gap he once estimated at two years has narrowed to six months. Schmidt also takes on the open source question that’s keeping Silicon Valley up at night, explains why kicking China’s best PhD students out of the country is one of America’s dumbest own goals, and shares the one thing he believes could keep this from becoming a zero-sum game — even if the current political environment makes it nearly impossible.
This conversation was recorded live as part of Masters of Scale Summit on October 22, 2026 at the Presidio Theatre in San Francisco.
Visit the Rapid Response website here: https://www.rapidresponseshow.com/
See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.
Google CEO Sundar Pichai: AI’s Infinite Possibilities, Compute Scarcity Dilemmas, Future AI Products
Sundar Pichai is the CEO of Google and Alphabet. Pichai joins Big Technology Podcast to discuss Google's return to the frontier of AI, its strategy for competing with OpenAI, and the enormous computing resources required to build the next generation of models. Tune in to hear why Google is sacrificing short-term Cloud revenue to support AI research, how demand for compute is stretching into 2027, and why its new Gemini 4 Argon model is seeing explosive adoption. We also cover recursive self-improvement, AI agents, the future of Google's products, the possibility of an AI bubble, data centers in space, and Google's partnerships with Apple and Anthropic. Hit play for an in-depth conversation about Google's AI strategy, the fierce competition shaping the industry, and why Pichai believes we're entering a moment of infinite possibilities.
---
Enjoying Big Technology Podcast? Please rate us five stars ⭐⭐⭐⭐⭐ in your podcast app of choice.
Want a discount for Big Technology on Substack + Discord? Here’s 25% off for the first year: https://www.bigtechnology.com/subscribe?coupon=0843016b
Learn more about your ad choices. Visit megaphone.fm/adchoices
Why every company now needs to think and operate like a lab team | Josh Woodward (Google Labs, Gemini, AI Studio)
Josh Woodward is the head of Google Labs, the Gemini app, and AI Studio. He has spent over 17 years at Google and oversees its longest-running and largest labs team, whose products include NotebookLM (now Gemini Notebook), Project Genie, and Flow. His job is to find, experiment with, and scale new AI products, and he has built one of the most distinctive product cultures in the industry for generating ideas, running experiments, and, even more importantly, knowing when to quit.
In our in-depth conversation, we discuss:
1. Why every company now needs to think and operate like a labs team
2. Where the best ideas really come from
3. The “almost possible” framework
4. How to recognize genuine product-market fit early
5. When to kill an idea
6. What skills are rising in value in the AI era
7. Why roles are not actually blurring into one universal builder role
8. How to set up your internal labs team
—
Brought to you by:
WorkOS—Make your app enterprise-ready, with SSO, SCIM, RBAC, and more
DX—Engineering intelligence for the AI era
—
Episode transcript: https://www.lennysnewsletter.com/p/why-every-company-now-needs-to-think
—
Archive of all Lenny's Podcast transcripts: https://www.dropbox.com/scl/fo/yxi4s2w998p1gvtpu4193/AMdNPR8AOw0lMklwtnC0TrQ?rlkey=j06x0nipoti519e0xgm23zsn9&st=ahz0fj11&dl=0
—
Highlights: https://lennyspodcast.com/mostreplayedmoments
—
Where to find Josh Woodward:
• X: https://x.com/joshwoodward
• LinkedIn: https://www.linkedin.com/in/joshwoodward
—
Where to find Lenny:
• Newsletter: https://www.lennysnewsletter.com
• X: https://twitter.com/lennysan
• LinkedIn: https://www.linkedin.com/in/lennyrachitsky/
—
In this episode, we cover:
(00:00) Introduction
(02:53) Why every company now needs to think like a labs team
(05:43) Where great ideas come from
(07:43) The “almost possible” framework
(11:52) Finding product-market fit
(16:09) Falling in love with the problem, not the solution
(17:57) When to kill an idea
(19:35) Overseeing Google Labs and the Gemini app
(22:21) Next consumer AI breakthroughs
(24:32) Google’s vision for a personal assistant
(27:18) What is overhyped and underhyped in AI right now
(31:36) Skills rising in value: unlearning rate, explosive endurance, trust
(35:45) Why roles are not blurring into one universal builder role
(38:59) Explosive endurance rhythms
(41:31) How planning has changed: rolling windows and 100-day milestones
(43:37) Labs lifecycle stages
(47:37) Tips for setting up an internal lab
(54:00) The future of products in an AI world
(56:26) Google Labs culture, rituals, and closing thoughts
—
References: https://www.lennysnewsletter.com/p/why-every-company-now-needs-to-think
—
Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email podcast@lennyrachitsky.com.
—
Lenny may be an investor in the companies discussed.
To hear more, visit www.lennysnewsletter.com
Google's AI Infrastructure Chief, Amin Vahdat, on the Physics & Economics of Frontier AI
At 100,000-accelerator scale, something fails multiple times an hour, which is why Google's Chief Technologist for AI Infrastructure Amin Vahdat thinks FLOPS is a vanity metric. The metric that matters is what he calls goodput: the useful work a workload actually delivers through real failures. Amin walks through the calculus that split the TPU line into 8i and 8t for the first time, why the TPU's core primitives haven't changed since v1, and how Google and DeepMind co-design in the same rooms, intercepting chip architectures mid-flight before tape-out. He explains why long-horizon agents are sending demand for CPUs and storage through the roof alongside accelerators, how optical circuit switches reroute light to a spare rack in milliseconds, and why Google would rather wait on a utility than build its own gigawatt. We also cover orbital data centers and the multi-megawatt rack of 2036.
Hosted by Sonya Huang, Sequoia Capital
0:00 – Introduction
1:47 – What makes a data center an AI data center
5:30 – Goodput, not FLOPS: holding yourself accountable when something fails every hour
11:52 – Doubling token capacity every six months, and where the gains actually come from
15:32 – The TPU bet: from a contrarian call in 2013 to splitting 8i and 8t
23:30 – The case for and against co-design
26:11 – Shoulder to shoulder with DeepMind: intercepting chips mid-flight
34:16 – Long-horizon agents change the shape of the data center
37:50 – Optical circuit switching and the state of networking
42:35 – Power is the binding constraint: utilities, gigawatts, and sizing a data center
49:23 – Training vs. serving clusters, seven-year-old TPUs, and open standards
58:29 – Orbital data centers and the supercomputer of 2036
🔬 An Oscar, Two Asteroids, and the Algorithm in Your sklearn: John Platt on AI for Science
<p>How often do you get to talk to a guest who has both an Academy Award and who invented textbook machine learning algorithms? John Platt has an Oscar, two textbook algorithms, two named asteroids, and an Erdos-Bacon number of 6. This was easily the most...</p><p><a href="https://www.latent.space/p/john-platt">Full show notes and links</a></p>
Inside Google’s Database Infrastructure for the AI Era
Historically, databases were responsible for storing data and returning exact results in response to queries. However, AI is now bending that contract in a new direction. Applications increasingly expect structured and unstructured data to come together. This is pushing databases into territory that looks more like search, where relevance and ranking matter and results are no longer strictly exact. Agents are also beginning to write their own queries and even propose their own schemas, which raises new questions about how data should be structured, governed, and trusted.
Sailesh Krishnamurthy is a VP of Engineering at Google, and in this episode he joins Matt Merrill to discuss his background, how databases have evolved over the past fifty years, and where the field is heading as AI reshapes how data is queried, structured, and trusted.
Sponsorship inquiries:
sponsor@softwareengineeringdaily.com
The post Inside Google’s Database Infrastructure for the AI Era appeared first on Software Engineering Daily.
732: Thomas Steiner on New Cross-Origin Storage APIs
Show Description
We're joined by Thomas Steiner from Google to discuss his proposal for cross-origin storage, a new web API designed to bring back safe, cross-site resource sharing. We explore why the old shared-CDN caching model collapsed due to privacy and fingerprinting risks, and how this could solve those issues while avoiding those same pitfalls.
Listen on Website
Watch on YouTube
Guests
Thomas Steiner
Guest's Main URL • Guest's Social
Developer Relations Engineer at Google, focused on the Web and Project Fugu.
Links
David Bushell
daverupert.com
633: Thomas Steiner on AI in Chrome and the Web
Pervasive Resources
Intent to ship: Cache sharing for extremely-pervasive resources
1037: WebMCP is here (and you should care)
Scott and Wes talk about WebMCP with Sarah Drasner and Dominic Farolino from the Chrome team, which is the new W3C standard that lets your site hand real tools to an agent. They cover the security model, headless and multi-tab agents, and how to wire it into an app you’ve already shipped.
Show Notes
00:00 Intro
00:46 Welcome to Syntax!
01:18 What Is WebMCP and How Is It Different from MCP?
02:32 WebMCP for UI: Enabling Agents to Use Websites Directly
06:05 Brought to you by Sentry!
06:43 Is This Just a Chrome Thing? W3C Standards and Open Development
11:49 WebMCP Security: Annotations, Hints, and Multi-Layered Defense
15:26 Best Use Cases: E-Commerce, Video Editing, and Complex UIs
16:27 The WebMCP Challenge: 5,000 Entries and Creative Applications
19:20 Headless WebMCP: Running Agents in the Background
22:43 Multi-Tab Scenarios: Using WebMCP Across Multiple Sites
24:23 WebMCP and Security: Safe Origin Policy and Agent Containment
28:09 Performance and Core Web Vitals for the Agentic Web
38:11 The Future of Web Monetization: Ads, Commerce, and Agent Payments
Universal Commerce Protocol
x402
40:53 Building an Agent Platform: Extensions, Harnesses, and Browser Vendor Roles
45:19 Support from Anthropic, Safari, Firefox, and Other Players
49:11 Getting Involved: Origin Trial, Community Feedback, and the W3C Spec
Ora
52:45 Sick Picks + Shameless Plugs
Sick Picks
Scott:
Wes:
Dominic: Steven Pinker: The Sense of Style
Sarah: Vintage Story
Shameless Plugs
Scott:
Wes:
Dominic: X account
Sarah: 25% discount with code Community25 at AGNTCon
Hit us up on Socials!
Syntax: X Instagram Tiktok LinkedIn Threads
Wes: X Instagram Tiktok LinkedIn Threads
Scott: X Instagram Tiktok LinkedIn Threads
Randy: X Instagram YouTube Threads
1024: In Case You Missed It in August 2026
In ICYMI Episode #1024, Jon Krohn tracks the gap between AI investment and AI return, from the technology side to the people side. Hear from Pete Johnson, Jerry Yurchisin, Priyanka Vergadia and Tristan Handy, discussing why four out of five organizations have the structures for AI success in place while only one in five sees the returns, which decisions should never be handed to a language model however confident it sounds, how to structure Claude skills so that your output stops being slop and why the semantic layer matters more, not less, now that analytics agents are the ones asking the questions.
Additional materials: www.superdatascience.com/1024
Interested in sponsoring a SuperDataScience Podcast episode? Email natalie@superdatascience.com for sponsorship information.
In this episode you will learn:
(00:56) Vector Search, Agentic Memory and Effective RAG
(09:20) Mathematical Optimization in the Agentic AI Era
(17:30) Anyone Can Write Code Now, So What Gets You Hired?
(27:14) How dbt Won Analytics Engineering