Starting July 8, 2026, Anthropic can require Claude Free, Pro, and Max subscribers to complete Claude identity verification — submitting a government-issued photo ID and a live facial scan processed by third-party partner Persona Identities. The check takes under five minutes. Users on Team, Enterprise, and API plans are not subject to the new requirement.
Anthropic published the identity verification policy around June 8, 2026, with enforcement beginning July 8 — a roughly 30-day notice window. The announcement became the #1 story on Hacker News on June 22 with nearly 800 points, driving significant user pushback across tech communities. The process is handled by Persona Identities, not Anthropic directly — a deliberate privacy architecture decision.
What Claude Identity Verification Collects: ID, Selfie, and Biometric Data
Claude identity verification is Anthropic’s new system requiring individual users to prove who they are before continuing to access the service. The policy is specific about what gets collected: a government-issued ID image, the information printed on that ID (including ID number and date of birth), a live selfie or short video, and facial geometry data. That last item may qualify as biometric data under laws in several US states and countries.
Anthropic says the data is used exclusively to confirm your identity. According to the official Claude Help Center policy, Persona is “contractually limited” to using the data for verification and fraud prevention — it cannot build profiles, share the data with third parties for marketing, or monetise it separately.
The verification window itself takes under five minutes on any device with a front-facing camera. Persona Identities was chosen by Anthropic for its “technology, privacy controls, and security safeguards” — language suggesting the company ran a competitive evaluation before selecting a partner.
Business Accounts Are Exempt — This Is a Consumer Policy
Team, Enterprise, and API customers face no verification requirement. The exemption makes the policy’s intent clear: this targets individual consumers, not the enterprise accounts that generate the bulk of Anthropic’s revenue. Business users typically operate under legal agreements, procurement reviews, and corporate IT oversight that already perform due diligence on who is accessing the service.
The consumer-only scope also has competitive implications. ChatGPT and Gemini currently impose no biometric verification requirements on their standard consumer plans. That makes Claude the first major general-purpose AI to cross this threshold at the individual subscriber tier. Whether competitors follow — or use this moment to position themselves as privacy-friendlier alternatives — will depend on how Anthropic’s churn numbers look after July 8.
Anthropic Won’t Say Who Gets Prompted — or What Triggers It
The policy language is deliberately ambiguous: Anthropic says it can require verification in certain situations — not that every Free or Pro subscriber will be prompted on July 8. The exact trigger conditions have not been disclosed publicly. There is no way to know whether you will be affected until a verification gate appears during a session.
Anthropic began testing the system in mid-April 2026 on select use cases before this broader rollout, according to Cybernews reporting. That timeline suggests July 8 is a hard enforcement deadline for full readiness, not a universal opt-in moment where every user faces a simultaneous prompt.
How This Fits Anthropic’s Broader Safety Posture
Anthropic frames the policy as a safety and fraud-prevention measure, consistent with its “responsible scaling” framework. Identity verification is one of the few tools that can definitively link a Claude session to a real person — useful for deterring high-risk misuse that is harder to detect algorithmically.
June 22 has been a turbulent day for Anthropic consumer services: a separate 90-minute outage at 00:37 UTC took down five Claude models simultaneously, and the same day saw Fable 5 — Claude’s most powerful model — shift from included on Pro/Max plans to usage-credit billing. The identity verification announcement compounds the sense of a rough week for individual Claude subscribers.
Frequently Asked Questions
Does every Claude user have to verify their identity?
Not necessarily. Anthropic says it can require Claude identity verification for Free, Pro, and Max plan users — but not that all of them will be prompted on July 8. The exact circumstances that trigger a verification check have not been publicly specified. Team, Enterprise, and API users are explicitly exempt.
What biometric data does Anthropic collect during verification?
Anthropic collects four data types: a government-issued ID image, the information on that ID (including ID number and date of birth), a live selfie or short video, and facial geometry data. The facial geometry component may qualify as biometric data under laws in certain US states and other jurisdictions.
Who processes the facial scan for Claude?
Persona Identities, a specialist third-party identity verification company, handles the entire process. Persona is contractually restricted from using collected data for any purpose other than verification and fraud prevention. Anthropic selected Persona for its technology, privacy controls, and security safeguards.
When does Claude identity verification take effect?
The policy takes effect July 8, 2026. Anthropic announced it around June 8, 2026 — giving affected users approximately 30 days of notice. The verification process takes under five minutes on any device with a camera.
What happens if I decline to complete identity verification?
Anthropic has not detailed the exact consequences of declining a prompt. Given the policy language, access to Claude could be suspended or restricted for users who are prompted and do not complete the check — but the company has not confirmed what refusal means in practice.
This is a developing story about one of the world’s most valuable AI companies. For context on recent Anthropic service changes, see our coverage of Claude usage limits and compute constraints and the Anthropic IPO filing.
Last Updated: June 2026
