Top AI Product

This isn’t a product. It’s a policy shot. Dario Amodei signed “Anthropic: Our position on open-weights models,” a blog post that hit Hacker News on July 28 and cleared 580 points in a few hours. It’s Anthropic explaining why the US should think hard about open-weight models — which, conveniently, is also a defense of Anthropic’s own closed-frontier bet.

What it actually says

Amodei denies the accusation that Anthropic wants to ban open models. His pitch: safe open-weight models are a public good, but instead of a ban, the US should choke China’s chip supply, crack down on large-scale distillation, and require mandatory safety testing on cyber, bio, and alignment risk before any capable model ships.

Why it’s blowing up now

Timing is everything. This lands as Kimi K3, Qwen, and DeepSeek — all open-weight, all Chinese — grab real US token share. Vance and the 2026 threat assessment worry about authoritarian governments piggybacking on open AI. Anthropic doesn’t open-source anything, so “be careful with open weights” reads as both a security argument and a moat.

You Might Also Like

• Ornith 1 0 Deepreinforce Self Scaffolding Coding Models Open Weights That Write Their own rl Scaffold 397b Hits 82 4 on swe Bench
• White House Accuses Moonshot ai of Distilling Anthropics Fable to Build Kimi k3 3 4 Million Claude Exchanges one Week Before the Weights Open
• Kimi k3s 2 8t Open Weights are Live but Running Them Takes 64 Gpus
• Deepseek v4 Just Dropped and the Open Source ai World is Losing its Mind
• Moonshine Open Weights stt the Tiny Speech Model That Punches way Above its Weight