China weighs curbing its top AI models; humanoid robots go public

China's Ministry of Commerce has held meetings with Alibaba, ByteDance, and startup Zhipu AI about whether to limit foreign access to the country's most capable models, according to Reuters — a striking reversal for labs whose open-weight releases have been the main challenge to US frontier dominance. Options sketched in the talks reportedly range from security reviews to barring the most sensitive models from public release, and cover open-weight systems like Qwen, Doubao, and GLM, not just proprietary ones. Nothing is decided, and officials have made no public comment.

Warehouse-robot maker Agility Robotics is going public through a special-purpose acquisition company, or SPAC, in a deal valuing it near two-and-a-half billion dollars and raising more than 620 million dollars — the largest capital raise in humanoid-robotics history. Its Digit robot has already logged tens of thousands of hours at customers like Toyota and GXO, and CEO Peggy Johnson pointedly put a home humanoid "ten-plus years" away. China's Unitree is pursuing a Shanghai listing in parallel. Public filings will force the industry to publish real deployment and margin numbers for the first time.

Security firm Wiz disclosed GhostApproval, a trust-boundary weakness found in six major AI coding assistants: Amazon Q Developer, Anthropic's Claude Code, Augment, Cursor, Google Antigravity, and Windsurf. It abuses decades-old symbolic-link behavior — a file inside a repository disguised as harmless settings can actually point at a developer's SSH keys, letting a malicious project write outside its sandbox and, in some cases, run code. Amazon, Google, and Cursor shipped fixes; Anthropic disputed that Claude Code's behavior is a bug, arguing a user who approves an edit in a trusted directory owns that call.

Anthropic named Ben Bernanke — who chaired the Federal Reserve through the 2008 financial crisis and won the 2022 Nobel Prize in Economics — to its Long-Term Benefit Trust, the independent body that can appoint members to Anthropic's board and advise leadership on AI's risks and societal impact. The company framed his role around one of its central research questions: how AI is reshaping the economy and the labor market. He joins three existing trustees as Anthropic moves toward a public listing.