scripod.com

Mira Murati's 975B Open Model, Ramin Hasani on Post-Transformer AI, and Demis' AI FINRA | EP #271

Shownote

The mates discuss Mira Murati’s 975B Open Model, Ramin Hasani speaks on Post-Transformer AI, and Demi’s AI FINRA. Get access to metatrends 10+ years before anyone else - https://qr.diamandis.com/metatrends   Peter H. Diamandis, MD, is the Founder of XPRI...

Highlights

In this episode, the panel dives into the latest frontier of AI, from a massive new open-weight model to a radical departure from transformer architecture. They debate the practicality of government regulation, the potential for AI to improve itself, and how biological inspiration is creating more efficient systems. The conversation also touches on the future of patents, AI's superior performance in medicine, and a surprising breakthrough in reversing a key marker of aging.
11:05
Regulatory capture could stifle open-weight models and innovation.
25:53
Open-weight models prioritize customization over leaderboard dominance.
50:07
Fine-tuning a 2 billion parameter model would take 350 years
1:15:25
A worm with 302 neurons inspired our liquid neural networks.
1:52:56
A previously permanent aging category becomes reversible.

Chapters

Is the government's plan to limit AI releases to China's level a recipe for disaster?
00:00
Mira Murati's 975-billion-parameter bet: Can an open model outpace the Chinese competition?
25:53
Why recursive self-improvement might be a dead end for today's AI models.
50:07
Worm brains in your car: How Liquid AI is building a post-transformer future.
1:10:15
From secret patents to reversing aging: The wildest breakthroughs in AI and longevity.
1:39:44

Transcript

Peter H. Diamandis: Mira Murati, the former OpenAI CTO, just shipped her first model. It's called Inkling. Customization over leaderboard dominance is what's going to win her the day. Salim Ismail: She's built exactly the thing hitting the market, that ex...