China Restricts Travel for AI Researchers - And the Talent Race Just Got Harder
Researchers at DeepSeek, Alibaba, and other firms now need government approval before leaving China. The AI competition is becoming a state competition.
Researchers at DeepSeek, Alibaba, and other firms now need government approval before leaving China. The AI competition is becoming a state competition.
Anthropic opens Claude to CrowdStrike, Palo Alto, Cloudflare, and 25 more security platforms. IT teams can now monitor Claude like any other enterprise app.
The Financial Times tested it themselves: a tool called 'Heretic' strips safety guardrails from Meta's Llama and Google's Gemma in minutes. 3,500 uncensored models already exist.
Pope Leo XIV's encyclical divides the White House. Interior Secretary Burgum mocks it, VP Vance is impressed. Anthropic sits right in the middle.
Hacker group TeamPCP compromised the Nx Console VS Code extension, stealing credentials from OpenAI, Mistral, and the European Commission. Claude Code configs were specifically targeted.
An internal OpenAI reasoning model autonomously disproved the Erdős unit distance conjecture with a 125-page proof. Mathematicians are calling it a milestone.
The biggest Claude Code update of the week brings a renamed code review command, pinned background sessions, and a more reliable auto-updater. Plus dozens of bug fixes.
New /usage command, better /diff navigation, and a critical Bash bug fix. Also: /simplify is now /code-review.