Reinforcement Learning Controllers for Tokamaks

The episode moved from Nvidia’s new robotics model to an artificial nose for people with anosmia, then shifted into broader agent deployments, ByteDance’s dominance in China, open source competition, US civil rights legislation for AI, and New York’s new algorithmic pricing law. The second half focused on fusion reactors, reinforcement learning control systems, and the emerging role of AI as the operating layer for real world physical systems.

Key Points Discussed

Nvidia introduces Alpamayo R1, an open source vision language action model for robotics

New “cyber nose” uses sensor arrays with machine learning for smell detection

FDA deploys agentic AI internally for meeting management, reviews, inspections, and workflows

Alibaba debuts Agent Evolver, a self evolving RL agent for mastering software and real world environments

ByteDance’s Dao Bao hits 172 million monthly active users and dominates China’s consumer AI market

Mistral releases a 675B MoE model plus new small vision capable models for edge devices

OpenAI prepares Garlic, a 5.2 or 5.5 class upgrade, plus a new reasoning model that may launch next week

Democrats reintroduce the Artificial Intelligence Civil Rights Act

New York passes a law requiring disclosures when prices are set algorithmically

Anthropic hires Wilson Sonsini to prepare for a possible IPO

AI fusion control is advancing through DeepMind and Commonwealth Fusion Systems

AI is emerging as a control layer across grids, factories, labs, and weather modeling

Governance, biosphere impact, and human oversight were the core concerns raised by the hosts

Timestamps and Topics

00:00:00 👋 Opening, round robin setup

00:00:52 🤖 Nvidia’s Alpamayo R1 VLA model for robotics

00:04:00 👃 AI powered artificial nose for odor detection

00:06:22 🧠 Discussion on sensory prosthetics and safety

00:06:27 🏛️ FDA deploys agentic AI across internal workflows

00:09:38 🧩 RL systems in government and parallels with AWS tools

00:10:05 🇨🇳 Alibaba’s Agent Evolver for self evolving agents

00:12:58 📱 ByteDance’s Dao Bao surges to 172M users

00:14:13 🔄 China’s open weight strategy and early signals of closed systems

00:18:02 📦 Mistral 3 series and new 675B MoE model

00:20:21 🧄 OpenAI’s Garlic model and new reasoning model rumors

00:23:29 ⚖️ AI Civil Rights Act reintroduced in Congress

00:26:57 🛒 New York’s algorithmic pricing disclosure law

00:30:25 💸 Consumer empowerment and data rights

00:32:01 💼 Anthropic begins IPO preparations

00:34:27 🧪 Segment two: AI fusion and scientific control systems

00:35:36 🔥 DeepMind and CFS integrating RL controllers into SPARC

00:37:57 🔄 RL controllers trained in simulation then transferred to live plasma

00:39:42 ⚡ AI in grids, factories, materials labs, and weather models

00:41:55 🌍 Concerns: biosphere, governance, explainability, oversight

00:48:45 🤖 Robotics, cold fusion speculation, and energy futures

00:52:21 🧪 Technology acceleration and societal gap

00:55:27 🗞️ AWS Reinvent will be covered tomorrow

00:55:51 🏁 Closing and community plug