Moonshots with Peter Diamandis

Tesla’s flying Roadster, collapsing costs for robotic intelligence, and AI tackling advanced physics anchor this wide-ranging conversation. Peter Diamandis, Salim Ismail, Emad Mostaque, and Anousheh Ansari connect breakthrough tech to incentives, XPRIZE competitions, AI alignment, and human purpose.

This episode centers on major AI and science headlines, from OpenAI solving 722 math problems at once to Anthropic’s policy on AI cruelty and consciousness. It also stretches into virtual cells, longevity, brain-computer interfaces, and Elon Musk’s vision for a quadrillion-dollar SpaceX.

AI safety takes center stage as Emad Mostaque joins the MOONSHOTS hosts to debate whether models like Claude could be conscious, why AI may be nearing a “1942 moment,” and what Sam Altman’s warning about accepting some bad outcomes really implies. The conversation also looks at the prospect of a Manhattan Project-style push toward superintelligence.

Astro Teller joins MOONSHOTS Live 2026 to unpack how X builds moonshots, why ambitious innovation is becoming more affordable, and how thousands of ideas are narrowed into breakthrough technologies. It’s a big-picture conversation about the systems behind bold bets on hard problems.

Self-improving AI is the focus here, as Richard Socher joins the hosts to unpack Recursive’s $670M bet, the path toward ASI, and why he assigns P(Doom) a zero. The conversation also touches on proposed limits for recursive self-improvement and what emerging benchmarks may reveal about AI’s next phase.