Claude can now ask for a passport or driver's license and a selfie

What's new

Anthropic has begun rolling out identity and business verification in Claude, but only for a few use cases. Users may be asked to prove who they are when using certain features, during routine platform integrity checks, or as part of other safety and compliance measures.

Verification is handled by the company Persona, which Anthropic has chosen as its provider. Users must show a government-issued photo ID, such as a passport, driver's license or national ID card, and take a selfie with a camera-equipped device. According to MediaNama, it applies to consumer accounts (Free, Pro and Max), not Team, Enterprise or developer platform accounts. No timeline has been given, and it is unclear exactly which actions trigger the requirement.

What's smart

The idea is to make Claude harder to abuse and to ensure that Anthropic knows who is using the more sensitive features. Anthropic says verification data is not used to train models, that only what is necessary is collected, and that the data is not shared for marketing or advertising.

Anthropic is the data controller, but it is Persona that stores the ID and selfie. Anthropic does not keep the images itself. According to Help Net Security, Persona is contractually allowed to use data only for verification and fraud prevention, and data is encrypted both in transit and at rest.

The requirements are strict: Photocopies, screenshots, scans and pictures of pictures are rejected. MediaNama reports that student IDs and temporary documents are also rejected.

Cheaper or better?

This is not a pricing story. No prices have been disclosed, and verification is not described as something users pay for. Anthropic's expanded privacy policy takes effect July 8, 2026, and describes, among other things, data shared with third-party integrations and verification requirements.

Compared with competitors, this could become a disadvantage. Help Net Security assesses that the move may prove unpopular and points out that, according to the article, ChatGPT and Gemini do not require similar verification. The Register reports that OpenAI uses age prediction rather than ID checks.

On Reddit and Hacker News, some users voiced dissatisfaction, and some considered canceling their subscriptions. They pointed to privacy concerns about Persona and its subprocessors. In February 2026, Discord drew criticism for a similar partnership with Persona and postponed its rollout.

One open issue is retention: Anthropic has not said how long verification data is kept. MediaNama also raises questions about how data protection rules fit with transferring data to the US, the legal basis and retention.

Who it's for

Verification is primarily aimed at people who use the features for which Anthropic wants extra security. For the vast majority of ordinary users, it is unlikely to change anything day to day, as long as they don't encounter a request.

If you get a request, you will need a valid, physical ID at hand and a camera. The sources do not make clear whether and when verification applies in Denmark and the EU, so it is worth watching before counting on it.