Claude introduces passport and selfie verification: full instructions for new Anthropic verification

On April 15, 2026, Anthropic quietly (that is, without a loud announcement) launched the **identity verification system for Claude. Now, to access certain features, register a Pro/Max subscription or as part of routine security checks, users are offered to undergo identity verification: download the original state document with a photo (passport, driver's license or national ID) and take a live selfie.

This is not a random experiment. Anthropic says, “Being responsible with powerful technologies starts with understanding who is using them.” Verification helps fight abuse, strictly comply with Usage Policy, age restrictions (18+) and legal requirements.

## Why Anthropic is taking this step now

Claude is one of the most “cautious” models of frontier AI. The company has long required proof of age at registration (18+), and in case of suspicion of a minor user, the account is blocked. Previously, Yoti was used for age verification, but there were false positives: users complained to Reddit that they were flagged as children even when using a VPN or flying.

The new system is the next level. It concerns:

• Expensive subscriptions (Max and Pro);
• Access to certain “hard” opportunities;
• Routine platform integrity checks;
• Cases where the account has come under suspicion.

Verification partner: Personal Identities. Anthropic chose them for “technical power, privacy controls and security measures.”.

## How exactly does the test take place: step by step

The process takes less than 5 minutes and is as simple as possible:

1. A Quick Identity Check or Identity Verification screen appears.
2. You need to have an original physical document with a photo on hand.
3. A smartphone camera or webcam takes a live selfie to compare a face.
4. The system automatically checks the data.

What is accepted:

• Passport;
• Driving license / state or provincial ID;
• National ID card.

What NOT to accept:

• Copies, scans, photos, screenshots;
• Digital/mobile versions of the ID;
• Student, corporate, bank cards and any non-governmental documents;
• Temporary paper IDs.

## Data Protection: What Anthropic Promises

Anthropic emphasizes that this is not data collection to train models. Here are the key guarantees:

• Anthropic is a data controller (they set the rules for storage and use).
• Images and data are not stored by Anthropic, but by Persona.
• Anthropic receives only verification confirmation, not the photos themselves (access is only possible on appeal).
• Everything is encrypted in transit and storage.
• Data ** is not used** for training models, is not shared with third parties for marketing or advertising.
• Removed in accordance with storage policy and legislation.

## What to do if verification fails

Most often, the problem is solved by a repeated attempt (best coverage, another document). If the attempts are over, fill out the support form on the Claude website.

Important: Even after successful verification, the account may be blocked for other reasons:

• Repeated violations of Usage Policy;
• Creating an account from an unsupported country;
• Violation of ToS;
• Confirmed use by minors.

In such cases, there is a separate form of appeal.

## User and market reactions

On X (Twitter) and Reddit, a wave of discussions immediately arose. Many call it “KYC for AI” and “digital border control.” Examples of real posts from April 15, 2026:

• “Claude requires a passport before subscription. ChatGPT and Gemini - No. Anthropic gave its competitors customers.”.
• Safety = treat every user like a suspect.
• Chinese users write that the screen appeared immediately when trying to open the Claude Max.

For Russian users, the situation is ambiguous. Claude is already officially unavailable in Russia: accounts are massively blocked, payment through intermediaries does not save. The new verification adds another barrier – especially if the account is already under suspicion due to IP or VPN. On the other hand, if you use a foreign number and a stable foreign IP, the chances of verification are high (passports of most countries are accepted).

## Comparison with competitors

## Results and what it means for users

Anthropic is betting on maximum liability and willingness to handle regulatory pressure. Frontier models are becoming too powerful to remain anonymous. This is a logical continuation of the trend: first age restrictions, then full KYC.

Plus:

• Reducing abuse and fake accounts;

• A safer environment;

• Compliance with future laws.

• Minuses:

• Barrier for new users;

• Privacy issues (even with all guarantees);

• Possible false locks.

If you plan to use Claude at a high level – get ready: passport at hand, good lighting and 5 minutes of time. For those who value anonymity above all else, competitors still look more attractive.