Claude puts up a wall as ID checks complicate access for Chinese users

What happened

Anthropic announced a new identity verification system for its AI assistant Claude that requires users to complete a real‑time selfie check while holding a government‑issued ID. The rollout was sudden and global, and it has been reported that the change is already visible to some international users. The check is meant to curb abuse and fraud, Anthropic says, but for many in China it amounts to an abrupt access barrier.

Why Chinese users are affected

Chinese users face a unique set of frictions. Domestic law and enforcement around personal data—alongside increasing caution about cross‑border transfers—make handing a national ID to a US company fraught. It has been reported that some Chinese users fear identity theft or regulatory exposure. And where many in China previously relied on VPNs or proxy services to reach foreign platforms, a real‑time ID selfie is a far higher hurdle. Expect some users to retreat to domestic alternatives such as Baidu (百度) and Alibaba (阿里巴巴), both of which have been accelerating local large language model offerings.

Broader context

Reportedly, Anthropic framed the move as part of broader safety and compliance work. But the timing and format intersect with geopolitics: export controls, sanctions, and China’s Personal Information Protection Law (PIPL) all shape how foreign AI services can operate and collect data. Is this an inevitable tightening of online identity, or an avoidable policy choice that shuts out millions of users? Either way, the update underlines a simple fact: verification and privacy rules that look reasonable in Silicon Valley can have outsized and unintended consequences inside China.