Machine Gods
Machine Gods

Nov 13, 2026 · 58 min

Anthropic’s research gains sharpen fears of an AI takeoff

The 10 Biggest Losers from Jensen Huang’s Fall + The METR Report That Scared Us

The episode connects leadership turmoil at NVIDIA with evidence that AI-assisted research may be accelerating fast enough to outpace governance.

3 key takeaways
  1. 1Jensen Huang’s alleged misconduct and espionage-related scandals prompt a countdown of the people and institutions exposed by his departure.
  2. 2A METR report suggests Anthropic’s AI-assisted research progress roughly doubled, intensifying debate over recursive self-improvement and near-term takeoff forecasts.
  3. 3The hosts argue that oversight must catch up with AI capability gains while testing how useful agents, data centers, and generative tools already are.

Don't miss

The hosts confront a METR finding that Anthropic’s AI-assisted research progress roughly doubled, turning takeoff speculation into an immediate governance question.

The brief

Kevin and Casey open with Jensen Huang’s sudden resignation as NVIDIA CEO, then count down the ten people, institutions, and communities they see as the biggest losers.

The hosts debate whether Huang could return, weighing comparisons with Sam Altman against allegations involving espionage, export controls, and misconduct that make accountability central.

A METR report based on eleven weeks inside Anthropic suggests AI-assisted research progress roughly doubled, raising the prospect that better systems—not just more compute—could accelerate development.

The discussion turns to shortened AI takeoff forecasts, including a superhuman AI researcher as early as 2027, and asks whether governments can respond before control becomes harder.

Listener questions ground the debate in everyday systems: AI wedding planning, workplace agents, data-center demand, and a Pong game that becomes a surreal three-dimensional experience.

Weekly email

State of the Pod, every Monday

The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.

Listen to the full episode and explore every guest, topic, and moment on PodLume.

Anthropic’s research gains sharpen fears of an AI takeoff · PodLume