Anthropic’s updated privacy policy β effective July 8 β says Claude may ask users to upload a government-issued ID, a selfie, and face geometry data to verify their identity. The Permission Layer just became personal.
Anthropic’s updated privacy policy, published in June and effective July 8, adds a new section: in “certain circumstances,” Claude will ask users to verify their age or identity by uploading a government-issued passport or driver’s license, a selfie photo or video, and a digitized face geometry template.
Anthropic says it uses Persona, a San Francisco-based identity verification company, to process the checks. Users may see verification prompts “when accessing certain capabilities, as part of routine platform integrity checks, or other safety and compliance measures.”
Anthropic spokesperson Michael Aciman clarified that this applies to a “small subset of users” whose accounts are flagged for potentially fraudulent activity β it’s an appeal mechanism, not a universal gate. Users who get flagged can verify instead of getting banned.
The context: This arrives two weeks after the US government ordered Anthropic to block Fable 5 and Mythos 5 from foreign nationals. The government’s letter didn’t specify how to verify nationality. Anthropic is now building the infrastructure to do exactly that β verify who is using its models, at the individual level.
The Structural Read
This is the Permission Layer moving from institutional to individual. Two weeks ago, access was gated at the organizational level β 200 Glasswing partners had Mythos, everyone else didn’t. Now Anthropic is building the plumbing to gate access at the person level.
FRONTIER AI IS BECOMING A CREDENTIALED SERVICE
No other softwareproduct asks for your passport. Not Google, not Microsoft, not Meta. Anthropic is treating frontier AI access like a regulated service β closer to opening a bank account or boarding an international flight than using a chat app. The implication: as models get more powerful, identity verification becomes the default, not the exception.
THE BIOMETRIC DATA QUESTION
Face geometry templates are classified as biometric data under Illinois’ BIPA law and similar regulations in Texas and Washington. Anthropic collecting facial geometry from Claude users opens a legal surface that no other AI company has touched. This is either forward-thinking compliance or a privacy liability β depending on how it’s implemented.
THE GOVERNMENT COMPLIANCE PLAY
The timing is not accidental. The US government told Anthropic to block foreign nationals. Anthropic is building the verification system to comply β and going beyond what was required by adding age verification and fraud detection. This positions Anthropic as the most compliant AI lab in the market. When GPT-5.6 launches, will OpenAI have the same infrastructure? Probably not.
The Bottom Line
Claude may want to see your ID. That sentence would have been absurd a year ago. Today it’s a privacy policy update from the company with the best AI model, the most government trust, and the infrastructure to enforce access at the individual level. The Permission Layer started as a theory about who gets to deploy frontier models. Now it’s a passport scanner built into a chatbot. The distance between “AI as a tool” and “AI as a controlled substance” just got shorter.
Business Engineer Framework
The Harness Society β When AI Access Requires a Passport
Gennaro is the creator of FourWeekMBA, which reached about four million business people, comprising C-level executives, investors, analysts, product managers, and aspiring digital entrepreneurs in 2022 alone | He is also Director of Sales for a high-tech scaleup in the AI Industry | In 2012, Gennaro earned an International MBA with emphasis on Corporate Finance and Business Strategy.
Scroll to Top
Discover more from FourWeekMBA
Subscribe now to keep reading and get access to the full archive.
