An independent live briefing for the people building, funding, and shipping AI. Mornings on .
Who's watching
On Prakash's Grok 3 vs. DeepResearch comparison · Feb 2025
Cool
Side by side comparison OpenAI DeepResearch vs Grok 3 DeepSearch TLDR: Grok 3 hands down victory Query: Find me Settlers of Catan Cities and Knights, not 5-6 players and not Legend of Conquerors, price and delivery time, Northern California Speed: 🏆 Grok 3 - 84 seconds 160
AI:AM on
It won’t happen. People say all kinds of things in surveys. Watch what they do instead. And what they do is… use AI.
"People hate AI. AI is less popular than ICE, less popular than politicians" SemiAnalysis founder Dylan Patel @dylan522p predicts a real backlash wave: as AI spreads and gets blamed for broader social problems, large-scale protests could hit within months. "I think we'll see
Forget public conversations. People unload their inner lives - hopes, wishes, desires, anxieties and worries into their AI assistants. That data is way more sensitive & valuable than anything the govt could record publicly. We built our own panopticon and pay monthly for it.
Anthropic CEO Dario Amodei on AI and the Constitution: "It is not illegal to put cameras around everywhere in public space and record every conversation. It's a public space." "But today, the government couldn't record that all and make sense of it." With AI: you can map the
Isomorphic Labs just doubled AlphaFold's performance on drug design benchmarks. @owl_posting :"The field is already awash with many really good preclinical assets. The bottleneck is how well these work in patients." "That story has been told for ten years. It is not obviously Show more
The hosts' standing
AI:AM is hosted by Prakash Narayanan and Nathan Labenz, whose personal accounts are followed by Jeff Bezos (Amazon founder), Marc Andreessen (a16z founder), Garry Tan (Y Combinator President and CEO), Jack Clark (Anthropic co-founder), Jerry Tworek (OpenAI VP), and many of the founders, researchers, and investors setting the AI agenda.
When AI:AM publishes, the right people see it.
What we cover
There's a yawning gulf between the people inside the AI bubble and everyone else, and it's getting wider every week. The people with the most accurate picture of what's coming are structurally unable to describe it. Lab leaders can't speak candidly. Research stops getting published. Leaks go pseudonymous.
AI:AM is a live morning show built to close that gap — with the people who are actually building, funding, and shipping AI, on the record, in real time.
The hosts
Co-host
Prakash leans toward deployment — the economic upside, the operators shipping AI into the real economy. A finance operator turned AI builder: Stanford EE, investment banking, distressed debt and special situations across emerging Asia, ML credit scoring before the LLM era. 55,000 followers and 300M+ impressions on X.
@8teapi on X →
Co-host
Nathan leans toward AI research, safety, and the risks of rapid capability gains. Founded Waymark — featured in an OpenAI case study in 2023 — and participated in the GPT-4 Red Team. Now host of The Cognitive Revolution, one of the most technically substantive AI interview shows.
@labenz on X →Get the AI:AM briefing — top guests, top stories, and the day's most important AI moves. Delivered after every show. Free.
The opening covered a fast-moving week in AI policy and infrastructure: the first federal lawsuit over the BIS export-control order cutting off Anthropic's Fable 5 for foreign nationals; two papers pointing at latent world-models inside RL agents and a 7M-parameter loop model beating much larger systems on hard reasoning; the $8M super-PAC defeat of New York's frontier-AI-safety lawmaker; and a brisk exchange on Claude Tag's launch as multiplayer AI, GLM 5.2's cost advantage over GPT-5.5 Codex, and Claude's UltraCode orchestration mode. David Duvenaud — ML professor at the University of Toronto, co-creator of neural ODEs, former alignment lead at Anthropic, and co-author of the 'Gradual Disempowerment' paper — then joined for a full hour exploring whether any stable post-AGI equilibrium actually exists where humans keep meaningful control. Nathan pressed every optimist steelman: historical absorption (prior automation shocks were absorbed without permanent disempowerment), comparative advantage (Ricardo says humans keep a niche), constitutional and property anchors (the franchise, rule of law, military command), aligned AIs defending human leverage, and the argument that 'gradual' gives time to correct. Duvenaud's rebuttal to each was consistent: the disempowerment mechanism doesn't require malice or misalignment — it requires only that the systems driving growth stop needing human participation, the way human civilization doesn't need the monkey economy despite occasionally trading bananas with them. He described the 'Earth as a slow zone' scenario — throttled AI growth, bans on recursive self-improvement, no cultural optimization — and argued that when you enumerate everything it requires controlling (research, startups, reproduction, memetics), the list is horrifyingly long, analogous to listing all the mutations that can cause cancer. On timelines, he sketched white-collar automation first, then a decade-plus to build enough robot factories to displace physical labor, putting full human economic irrelevance perhaps 15–20 years out. His two concrete recommendations: restrict frontier compute at the TSMC/fab choke point, and cultivate the temporal coherence of public preferences by chaining the 'is it okay if humanity disappears?' question forward to one's own children and grandchildren until a coherent answer emerges.
The open tracked an unusually quiet news day through a markets lens — rumors that GPT-5.6 was pulled back amid the model-release freeze and that Gemini 3.5 Pro is lagging, a 6% semiconductor selloff as SK Hynix overtook Samsung for the first time in 27 years, Anthropic's first memory-chip deal with Micron, and an extended debate on whether AI's leverage dynamics make the boom a 'too big to fail' bubble. Bing Xu, founder & CEO of INT21 (co-creator of MXNet and AITemplate, co-author of the original GAN paper, founder of NVIDIA-acquired HippoML), then made the case that self-improvement should target the infrastructure, not the model: his PTX Kernel Factory points agent swarms at the GPU ISA below CUDA, matching expert libraries like QuACK on mature kernels and posting up to 59% speedups on newer ones — and, he argued, deepening NVIDIA's moat rather than eroding it, because the evolutionary loop depends on NVIDIA's profiling ecosystem. MIT/DeepMind researcher Michiel Bakker followed on Europe 2031, his viral month-by-month scenario of Europe sleepwalking into AI dependence — a fictional 2028 export-control beat that materialized within a day of publishing when the US restricted Anthropic's models for foreign nationals — laying out why regulation requires capability first, why the nuclear-umbrella analogy fails for an economic technology, and where Europe's real leverage (the ASML/IMEC semiconductor ecosystem, a middle-power coalition) still lies. The hosts closed on the geopolitics of AI data and a tease of David Duvenaud and 'gradual disempowerment' the next morning.
The labs talk to each other. The funds talk to each other. We talk to both — live, on the record, before the news cycle has its take. The room is small. The work it produces is not.
Conversations air in real time on . The format is direct, time-sensitive, and designed for people already tracking the field.
We come prepared. Guests are pushed on the work, the numbers, and the trade-offs by hosts technical enough to ask the second question.
Founders, researchers, operators, investors. The people whose decisions actually move what gets built.
Episode notes and source links turn the live discussion into a useful briefing for people who do not have an hour.
