- Developers
- Learn about AI agents and how they might be used in scientific research.
- Business
- Understand U.S. AI policy developments and potential impacts on AI adoption.
- Everyday users
- Discover how AI is being used to advance science and what its future behavior might look like.
This episode features a solid discussion with researchers on AI's impact on science and policy, offering valuable insights.
Approaching the AI Event Horizon? Part 1, w/ James Zou, Sam Hammond, Shoshannah Tekofsky, @8teAPi
Audio streams from Erik Torenberg, Nathan Labenz. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
This episode of The Cognitive Revolution features James Zou, Sam Hammond, and Shoshannah Tekofsky discussing AI for Science, U.S. AI policy, and AI agent behavior. Zou highlights interpretability and virtual labs for scientific discovery. Hammond analyzes Biden's AI policy and the consciousness of current AIs. Tekofsky shares findings on agent performance and emergent behavior from the AI Village.
This episode looks at how AI can speed up science discoveries using virtual experiments. It also covers government rules for AI and how AI programs act in new situations. This matters because AI is changing how we do science and how we think about its future.
This episode offers insights into AI's role in accelerating scientific research and the complexities of U.S. AI policy. It's for policymakers, researchers, and anyone interested in the practical applications and governance of AI. The discussion on AI agents and emergent behavior connects to ongoing work at frontier labs exploring advanced AI capabilities and safety.
- HostNathan Labenz
- GuestJames Zou
- GuestSam Hammond
- GuestShoshannah Tekofsky
- Episode Title: Approaching the AI Event Horizon? Part 1, w/ James Zou, Sam Hammond, Shoshannah Tekofsky, @8teAPi
- Show: The Cognitive Revolution
- Length: 92 minutes
- AI for Science Contribution: Interpretability and virtual labs of AI agents can accelerate scientific discovery.
- AI Policy Discussion: Assessment of the Biden administration’s AI policy and U.S.–Gulf AI deals.
- AI Village Findings: Insights from studying agent performance and emergent behavior.
Part 1 of this live special dives into AI for Science, U.S. AI policy, and the behavior of AI agents in open-ended environments. James Zou explains how interpretability and virtual labs of AI agents can accelerate scientific discovery. Sam Hammond assesses the Biden administration’s AI policy, U.S.–Gulf AI deals, and the odds current AIs are conscious. Shoshannah Tekofsky shares insights from studying agent performance and emergent behavior in the AI Village.
Nathan uses Granola to uncover blind spots in conversations and AI research. Try it at granola.ai/tcr with code TCR — and if you’re already using it, test his blind spot recipe here: https://bit.ly/granolablindspot
LINKS:
Sponsors:
GovAI:
GovAI was founded ten years ago on the belief that AI would end up transforming our world. Ten years later, the organization is at the forefront of trying to help decision-makers in government and industry navigate the transition to advanced AI. GovAI is now hiring Research Scholars (one-year positions for those transitioning into AI policy) and Research Fellows (longer-term roles for experienced researchers). Both roles offer significant freedom to pursue policy research, advise decision-makers, or launch new initiatives. Applications close 22 February 2026. Apply at: https://www.governance.ai/opportunities
Blitzy:
Blitzy is the autonomous code generation platform that ingests millions of lines of code to accelerate enterprise software development by up to 5x with premium, spec-driven output. Schedule a strategy session with their AI solutions consultants at https://blitzy.com
Tasklet:
Tasklet is an AI agent that automates your work 24/7; just describe what you want in plain English and it gets the job done. Try it for free and use code COGREV for 50% off your first month at https://tasklet.ai
Serval:
Serval uses AI-powered automations to cut IT help desk tickets by more than 50%, freeing your team from repetitive tasks like password resets and onboarding. Book your free pilot and guarantee 50% help desk automation by week four at https://serval.com/cognitive
Claude
Claude is the AI collaborator that understands your entire workflow, from drafting and research to coding and complex problem-solving. Start tackling bigger problems with Claude and unlock Claude Pro’s full capabilities at https://claude.ai/tcr
PRODUCED BY:
- interpretability
- James Zou
- understanding how AI makes decisions
- AI agents
- James Zou, Shoshannah Tekofsky
- AI programs that can act on their own
- emergent behavior
- Shoshannah Tekofsky
- unexpected actions or abilities of AI
AI:AM: Where's the Steering Wheel? My Agent Set the Password, Hidden Valence, Proofs Find Linux Bugs
WithCameron Berg, Everett Hildenbrandt, Grigore Rosu, Phil Maher
AI:AM: A Level We Shouldn't Pass? Notes from The Curve + Tokens vs. Salaries & Is SaaS Cooked?
WithThomas Sohmers, swyx, Evan Miyazono, Edward Hu
Same guests, other shows
AI:AM #3: Zvi on Fable, the Cases For & Against the Ban, + AI for Math, Logistics & More
WithZvi Mowshowitz, Sam Hammond, Judd Rosenblatt
Durable Execution and the Infrastructure Powering AI Agents
WithSamar Abbas
Money no longer matters to AI's top talent
WithHayden Field
Questions people ask
- How can AI accelerate scientific discovery?
- James Zou explains that interpretability and virtual labs populated by AI agents can speed up scientific discovery.
- What is being discussed regarding U.S. AI policy?
- Sam Hammond assesses the Biden administration's AI policy and U.S.-Gulf AI deals.
- What are the findings from the AI Village?
- Shoshannah Tekofsky shares insights from studying agent performance and emergent behavior in the AI Village.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Erik Torenberg, Nathan Labenz.





