HomeNews

Anthropic Cuts Off Claude Access Across Mainland China

Rishi
Rishi
•
Nov 12, 2026 8:10 AM
•
0
 min read
Select Emergent as your Preferred news source
Anthropic Cuts Off Claude Access Across Mainland China

💡 TL;DR

  • Anthropic blocks Claude access from mainland China effective November 12, 2026, enforcing regional usage limits announced in October.
  • Users flagged for risk signals must submit government ID and selfie verification through Persona or face account closure.
  • OpenAI introduces similar controls starting November 20, requiring platform approval and limiting multi-account paid subscriptions.

Anthropic has cut off access to Claude across mainland China and other restricted regions, with the updated regional usage policy taking effect Thursday, November 12, 2026. The restrictions apply to Claude.ai, the Claude API, and cloud-distributed versions of the model, marking a significant tightening of geographic controls for one of the industry's leading AI platforms.

Policy Scope and Enforcement

The new rules prohibit Claude use from mainland China, by companies registered in restricted regions, and by any entity directly or indirectly controlled by individuals or institutions in those areas. Anthropic first announced the policy change on October 8, 2026, giving users and organizations approximately one month to adjust before enforcement began.

Risk signals that can trigger verification include unauthorized resale of Claude access and use from restricted regions. When flagged, users must complete identity checks through Persona, a third-party verification service. The process requires submission of a government-issued ID and a selfie. Failed verification attempts result in immediate account closure.

Identity Verification Requirements

Anthropic began rolling out identity checks in April 2026, establishing the infrastructure now being used to enforce regional restrictions. The Persona integration allows the company to verify user identity and location at scale, creating a technical barrier to access from prohibited regions.

Users attempting to access Claude from mainland China or through VPNs masking their location face immediate verification prompts. The system flags accounts showing inconsistent geographic signals or patterns suggesting resale activity. Once an account enters the verification queue, users have limited time to provide valid documentation before losing access.

Parallel OpenAI Policy Changes

OpenAI announced similar controls on October 21, 2026, with an effective date of November 20. Third-party platforms offering direct access to GPT models must now obtain advance approval from OpenAI. The company is also limiting use across multiple paid accounts, with payment cards tied to too many subscriptions blocked from purchasing new plans.

OpenAI's restrictions cover ChatGPT Go, Plus, and Pro plans. Eligible Pro subscribers can receive API credits equal to their subscription fee, capped at $500 monthly. The company prohibits account sharing and credit reselling, with violations leading to account closure, mirroring Anthropic's enforcement approach.

Release Date and Availability

Officially launched on November 12, 2026, Anthropic's regional access controls are now fully operational across all Claude services. The policy applies to existing accounts and new signups, with no grandfather period for users who previously accessed Claude from restricted regions. OpenAI's platform approval requirement and multi-account limits will officially launch on November 20, 2026.

What This Means

The simultaneous tightening of access controls by both Anthropic and OpenAI signals a coordinated industry shift toward stricter geographic and usage enforcement. For developers and organizations operating in or with connections to mainland China, the changes force immediate migration to alternative AI platforms or costly restructuring of corporate ownership and operations. The verification requirements also raise privacy concerns for users who must submit biometric data to maintain access, creating friction that may push some toward open-source alternatives. As leading AI labs implement these restrictions, the global AI market is fragmenting along geographic lines, with implications for research collaboration and commercial deployment across borders.

About the writer

Rishi drives Product and Growth at Emergent, bringing 11+ years of experience building and scaling products across fintech and consumer platforms. He previously co-founded Truly Rural and was Head of Product at Nigeria Fintech. Rishi holds an MBA from IIM Indore.

Start Building
on Emergent today
Try Emergent