Descriptions:
Peter Diamandis hosts Episode 260 of Moonshots with co-hosts Alex and Dave, opening with Anthropic’s Claude Opus 4.8 reclaiming the top position on the Artificial Analysis Intelligence Index at 61.4 — 1.2 points ahead of GPT 5.5. On SWEBench Pro, the hard coding benchmark, Opus 4.8 scored 69.2% versus GPT 5.5’s 58.6%, and it is the only model to complete every case end-to-end on Anthropic’s internal super-agent benchmark. Notably, Opus 4.8 is described as four times less likely to overlook bugs in its own code. The hosts frame the current moment as a two-horse race between Anthropic and OpenAI, with major releases arriving every four to six weeks.
The episode also covers OpenAI’s newly disclosed $220 billion nonprofit foundation — the largest philanthropic war chest in history — with a mandate focused on global stability through research into public wealth funds, worker ownership models, and AI dividends (UBI, universal basic compute, and universal basic equity). The hosts debate whether this structure resembles privatized socialism and what it signals about frontier labs becoming societal backstops.
Demis Hassabis of Google DeepMind tightened his AGI prediction to 2029, aligning with Ray Kurzweil’s longstanding forecast — though the panelists push back on whether the AGI framing itself remains useful. Additional topics include Amazon’s new AI shopping assistant, a Chinese pocket-sized cancer detector claiming 95% accuracy from a single blood drop, and a $2 billion US government and IBM investment to build a quantum chip foundry.
📺 Source: Peter H. Diamandis · Published June 01, 2026
🏷️ Format: Podcast
