The Government Is Deciding Which AI Model You Can Use

OpenAI just dropped GPT-5.6 in three tiers - Sol, Terra, and Luna - and yes, the naming scheme looks awfully familiar. This week, Pierce and Rich break down what the new lineup actually means, why nobody can prove when a model is "done" training, and how AI is getting scary good at chaining real exploits. Then things get spicy: Anthropic accuses DeepSeek, Moonshot, and Minimax of running 24,000 fake accounts and 16 million conversations to distill Claude. But is it hypocritical to cry theft when your own models were trained on the open internet? Plus a four-year tour of US AI policy, from the AI Bill of Rights to pre-release cyber reviews. --- 00:00 Intro 01:12 GPT-5.6 Naming Tiers and AI Cyber Capabilities: Harnesses, Exploit Chains, and AISI Benchmarking 31:22 US Government AI Policy Timeline: From the NIST Risk Framework to Cyber Pre-Release Review 48:50 Anthropic Names Chinese AI Labs: Distillation, Stolen Thinking Tokens, and the Copyright Hypocrisy 01:02:37 Outro 01:02:59 Post Show Banter