Back to News

AI researchers discuss case against RSI, Chinese labs' progress, long-horizon RL on Dwarkesh Patel podcast

#podcast#ai-research#rsi#reinforcement-learning

Dwarkesh Patel's podcast released a Q&A episode featuring AI researchers John Schulman, Beren Millidge, and Charlie O'Neill. They discussed steelmanning the case against recursive self-improvement (RSI), progress by Chinese labs, and long-horizon reinforcement learning, with Schulman noting 'We're nowhere near the ceiling.'

Coverage timeline

  1. Techmeme
    Q&A with AI researchers John Schulman, Beren Millidge, and Charlie O'Neill on steelmanning the case against RSI, Chinese labs' progress, long-horizon RL, more (Dwarkesh Patel/Dwarkesh Podcast)

    Dwarkesh Patel / Dwarkesh Podcast : Q&A with AI researchers John Schulman, Beren Millidge, and Charlie O'Neill on steelmanning the case against RSI, Chinese labs' progress, long-horizon RL, more — “We're nowhere near the ceiling.” — New episode with John Schulman, Beren Millidge and Charlie O'Neill.

Related stories

  • Podcast: Epoch AI's JS Denain debates RSI, US-China gap, distillation, open vs closed safety

    An Interconnects podcast episode features a debate with Jean-Stanislas 'JS' Denain of Epoch AI, covering predictions for recursive self-improvement (RSI), the role of robotics in AI acceleration, the US-China AI gap, whether distillation explains the gap, insights from Chinese job postings, and whether open or closed models are safer. Both participants express significant uncertainty about the trajectory of AI.

    1 reports · 1 sources

  • OpenAI's Noam Brown Discusses Multi-Agent Systems and Navier-Stokes in Podcast

    In a Dwarkesh Podcast episode, OpenAI researcher Noam Brown discussed multi-agent systems, AI solving the Navier-Stokes problem, and the gap between internal and external models. Brown emphasized the importance of not underestimating AI capabilities.

    1 reports · 1 sources

  • Podcast: Redwood Research's Greenblatt debates AI R&D, RSI, data bottlenecks

    Dwarkesh Patel's podcast features an interview with Redwood Research Chief Scientist Ryan Greenblatt, covering topics including AI research and development, recursive self-improvement, whether human expert data is a bottleneck, token prices, and alignment. The conversation is framed as a debate about recursive self-improvement.

    1 reports · 1 sources

  • SemiAnalysis founder discusses $11T AI capex, compute centralization, China's role

    In a podcast interview, SemiAnalysis founder Dylan Patel discussed Anthropic and OpenAI's control over global compute, projecting $11 trillion in AI capital expenditure between 2024 and 2029, and addressed China's compute capabilities. Patel characterized the industry trend as 'every force is screeching towards centralization.'

    1 reports · 1 sources

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe