Anthropic has begun rolling out mandatory identity verification for some users of its AI assistant, Claude. According to an official support document published by the company, users may be required to submit a government-issued ID and a selfie to gain or maintain access, as part of an effort to prevent misuse of its Acceptable Use Policy — particularly around election-related misinformation.
A New Precedent for AI Accountability
Anthropic says the verification is designed to proactively deter the creation of policy-violating content. As AI models become more capable of generating convincing text and images, tying account access to real-world identity marks a notable shift from the largely anonymous access common across major AI platforms.
How the Verification Process Works
Anthropic is partnering with a third-party identity verification provider to handle the sensitive data. The process is being rolled out gradually, so not all users are being prompted to verify their identity at once. According to Anthropic's support documentation, the process works as follows:
- ID Submission: Users are prompted to upload a photo of a valid, government-issued photo ID.
- Live Selfie: A liveness check confirms the person submitting the ID is its legitimate owner.
- Data Handling: A third-party vendor processes the ID and selfie and confirms the user's identity to Anthropic.
- Confirmation, Not Data: Anthropic states it does not receive or store the ID photos or biometric data itself — only a confirmation that the check was passed.
Why It Matters
The policy represents a direct trade-off between user anonymity and platform safety, a dilemma facing every major AI lab. It could curb misuse of Claude, but it also introduces friction, raises privacy concerns, and may create accessibility barriers for people without government-issued identification. How rivals such as OpenAI and Google respond could shape the industry's broader approach to access and accountability.