Summary
The Moonshots podcast with Peter Diamandis and a five-person panel dissects one of the most viral AI safety news cycles in recent memory: the public resignation of Jacob Coxin, a researcher who spent three years at OpenAI and Anthropic, posting that neither company “is acting responsibly” in the race toward self-improving superintelligence. The post reached an estimated 100–150 million views, amplified by Anthropic alignment lead Evan Hubinger, who publicly stated he believes there is over a 10% chance AI could kill all humans within a decade — a combined reach of roughly 200 million impressions that crossed over to the nightly news within 24 hours.
The panel debates whether this represents genuine whistleblowing or an orchestrated campaign by EA-aligned doomer policy groups, with panelists pointing to the account’s blank history and the speed of coordinated amplification. The episode also covers OpenAI’s reported solution to the Navier-Stokes equations using 10,000 agents — which Sam Altman cited as evidence for urgency around safety pacing — the first AI-designed longevity drug, Google’s human genome mutation mapping initiative, and political reactions from Bernie Sanders and Illinois Governor JB Pritzker calling for AI regulation.
Across 18 stories, the through-line is whether humanity has entered a phase transition in AI progress — with machines solving century-old scientific problems faster than anyone expected — and what the implications are for safety, governance, and human relevance.
📺 Source: Peter H. Diamandis · Published September 11, 2026
🏷️ Format: Podcast
