Anthropic has started requiring government-issued identification from a subset of users on its Claude AI platform, linking access to certain subscriptions, advanced features, and account reviews to identity verification. The policy update surfaced in the company’s help documentation during the week of April 14–16, 2026, and appears to apply only in selected cases rather than across the full user base.
A targeted verification system, not a platform-wide rule
According to the company’s explanation, the new requirement is triggered in specific scenarios, including higher-tier plans, access to more advanced capabilities, or internal safety and integrity reviews. Anthropic has framed the rollout as a limited control measure designed to reduce abuse, enforce platform rules, and satisfy legal obligations, rather than a universal onboarding requirement for everyone who wants to use Claude.
Users who receive the verification prompt must submit a physical government-issued photo ID and complete a live selfie scan. Anthropic says the process typically takes less than five minutes and requires a device with a working camera. Accepted documents include passports, driver’s licenses, and national identity cards. The company explicitly rejects screenshots, digital copies, temporary paper IDs, and non-government credentials such as student or employee cards.
Persona handles the workflow as privacy concerns rise
The identity-check process is being managed by Persona, a third-party identity verification provider. Anthropic says the underlying ID images are not stored on its own systems. Instead, Persona retains the data under contractual limits, while Anthropic keeps access to the verification outcome when needed for account review, compliance checks, or appeals.
Anthropic also says the information is encrypted and used only for identity confirmation, fraud prevention, and compliance purposes. The company states that identity data is not used to train its AI models and is not shared for marketing. Disclosure, according to the firm, is limited to legal requirements. These assurances are likely aimed at reducing concerns from users who see identity-linked AI access as a significant privacy shift.
The company’s rationale reflects a wider industry challenge. AI platforms are under growing pressure to address misuse, including fraud, impersonation, and other harmful behavior. Anthropic has also pointed to age restrictions as part of the equation, with some accounts reportedly associated with users under 18 being suspended pending verification.
User backlash centers on privacy and competitive pressure
Reaction online has been largely negative. Critics on social media argued that Claude now asks some new subscribers to provide a government ID and photo through Persona, while competing products such as ChatGPT and Google Gemini do not currently require similar verification for ordinary chatbot access. That comparison has become central to the backlash.
Some users interpreted the move as an early version of “AI KYC,” warning that identity checks could become normalized as generative AI systems grow more powerful and more tightly scrutinized. Others complained that Anthropic may be introducing additional friction at a time when convenience, privacy, and ease of access remain major competitive factors in the AI market. On online forums, frustration was also directed at the broader possibility that future AI systems could become harder to use anonymously.
The competitive concern is straightforward: if one provider adds identity verification while others do not, users who value privacy may shift to less restrictive alternatives. The article notes that OpenAI and Google’s Gemini currently lack comparable requirements for standard consumer chatbot use. It also points out that some alternatives emphasize privacy or local-model usage, giving dissatisfied users more options if they want to avoid submitting official identification.
A possible sign of where consumer AI is heading
Although Anthropic’s system is currently selective rather than universal, its introduction may signal a broader policy direction for the AI sector. As platforms expand access to more capable tools, the tension between safety, compliance, abuse prevention, and user privacy is becoming harder to ignore. Identity verification may increasingly be presented by companies as a practical safeguard, especially in high-risk or premium-use cases.
At the same time, the company’s move highlights a difficult trade-off. Stronger verification can help deter fraud and make enforcement easier, but it can also raise barriers for legitimate users, intensify privacy concerns, and create a perception that AI services are moving toward full real-name tracking. In a market where access friction matters, even a limited verification policy can reshape how users compare platforms.
For now, Anthropic has not imposed government ID checks on every Claude user. But the existence of the system is significant in itself. It suggests that identity-based controls may become a more common layer in consumer AI, especially as regulatory pressure rises and providers face growing scrutiny over how their tools are used. Whether that becomes an industry standard or a competitive disadvantage will depend in large part on how rivals respond—and how much friction users are willing to tolerate in exchange for access to advanced AI capabilities.