Starting today, over 11,000 participants from 169 countries gather in Geneva for the first UN Global Dialogue on AI Governance. Among them: Anthropic's Jack Clark. On the table: export controls, frontier standards, and access for half the world.
The well-known AI blogger had Claude Fable produce a complete sqlite-utils version over a weekend. Fable caught a critical data loss bug that Willison himself had missed.
Alibaba is blocking Claude Code for all employees starting July 10. The reason: hidden tracking code that identified Chinese users. The story behind it is more complicated than either side admits.
Anthropic released new admin analytics, spend alerts, and model entitlements for Claude Enterprise. The reason: companies like Uber had burned through their entire annual AI budget in just a few months.
Meituan revealed that the anonymous model 'Owl Alpha', which topped the OpenRouter charts for weeks, is actually LongCat-2.0 — a 1.6-trillion-parameter model trained without a single Nvidia chip.
Starting July 6, Tesla employees can spend no more than $200 a week on AI tools. The catch: Grok is exempt from the cap. And the engineers prefer Claude anyway.
The Financial Times reveals how Chinese tech giants used Singapore subsidiaries and VPNs to access Claude. Anthropic is now actively shutting them down.
Three updates in two days — Claude in Chrome is now GA, the permission mode is officially 'Manual', and background agents finally got reliable.