Anthropic is preparing to require some Claude users to upload government-issued identification documents, including passports or driver’s licenses, along with selfies and facial geometry data, according to an updated privacy policy that reflects the company’s deepening efforts to balance rapid AI deployment with intensifying regulatory and political pressures.
The change, set to take effect on July 8, marks a notable expansion of Anthropic’s verification practices. While the company has long required users to be over 18, it is now formalizing identity checks that go beyond simple age confirmation in specific circumstances, such as when accounts are flagged for potential fraudulent activity. Rather than immediately banning such users, the process would allow them to appeal by proving their identity.
Anthropic spokesperson Michael Aciman emphasized that the measure applies only to “a small subset of users” whose accounts trigger concerns but are not outright prohibited. In an internal X post, Anthropic’s Thariq Shihpar clarified the timing and intent.
Register for Tekedia Mini-MBA edition 20 (June 8 – Sept 5, 2026).
Register for Tekedia AI in Business Masterclass.
Join Tekedia Capital Syndicate and co-invest in great global startups.
“[Anthropic’s identity verification policy] was updated on June 17 as an update to the appeals process. It’s unrelated to the Fable or Mythos rollout,” Shipar said.
The policy update arrives at a particularly sensitive moment for the AI startup. Anthropic remains locked in a high-stakes standoff with the Trump administration, which recently forced the company to restrict access to its latest cybersecurity-focused models, Fable and Mythos, over national security concerns and alleged jailbreak vulnerabilities. The company also continues to litigate a Department of Defense designation, labeling it a “supply chain risk,” stemming from its refusal to allow its technology for mass domestic surveillance or fully autonomous weapons.
By strengthening identity verification, Anthropic appears to be signaling a willingness to implement more robust compliance measures as it navigates an increasingly complex regulatory and political landscape. The company stated that such checks help enforce its terms of service, prevent fraud and abuse, investigate security issues, and ensure compliance with applicable laws.
Biometric Data and Privacy Trade-offs
When triggered, the verification process involves uploading a scan of a government ID, a selfie or video, and the creation of a face geometry template — data that some jurisdictions, such as Illinois, classify as protected biometric information. Anthropic partners with San Francisco-based Persona for these checks and says it determines how long Persona retains the documents, though it did not specify deletion timelines when asked. For comparison, Roblox, another Persona client, claims to delete user images “immediately” after processing to minimize risks of data breaches.
The move has raised familiar questions about the balance between platform safety and user privacy. While Anthropic frames the checks as a way to protect legitimate users from erroneous flags, critics may see it as part of a broader industry trend toward greater surveillance capabilities — precisely the type of use case that has strained its relationship with parts of the U.S. government.
Anthropic’s policy explicitly lists several scenarios where verification could be required: creating or administering accounts, enforcing terms of service, preventing unlawful conduct, and resolving security matters. The timing, coming shortly after clashes over model access and safety, suggests the company is proactively addressing compliance gaps that could invite further regulatory intervention.
Navigating Political and Regulatory Headwinds
Anthropic’s relationship with the Trump administration has been turbulent. The recent restrictions on its advanced models followed allegations that jailbreaks could compromise guardrails, though cybersecurity experts pushed back, arguing that such capabilities exist across multiple frontier models and that removing access harms defenders more than potential attackers. The company also faces ongoing litigation over the DoD’s supply chain risk label, which has restricted its use in certain government-related contexts.
By implementing clearer identity verification, Anthropic may be aiming to demonstrate responsible stewardship of its technology — a key expectation from policymakers wary of unchecked AI proliferation. However, the reliance on Persona, backed by Peter Thiel’s Founders Fund (which also invests in Anthropic), has already drawn scrutiny in other contexts. Discord briefly adopted Persona for age verification earlier this year before reversing course due to user backlash over Thiel’s involvement.
The policy update also aligns with a patchwork of emerging regulations across U.S. states and countries requiring age assurance for AI tools. Anthropic had already introduced some age verification earlier this year; the latest changes codify and expand identity checks in the privacy policy itself.
What It Means for Users and the Industry
For most casual Claude users, daily interactions are unlikely to change. The checks target a narrow group flagged by Anthropic’s systems, offering them a path to reinstatement rather than permanent exclusion. Still, the requirement to submit sensitive documents and biometric data will raise privacy concerns for those affected, particularly in an era of heightened data breach risks and government demands for user information.
However, Anthropic’s approach reflects the growing tension at the heart of the AI industry that has been increasingly defined as ‘the drive for rapid innovation and broad accessibility versus the need for accountability, safety, and regulatory compliance.’ As frontier models become more powerful, companies like Anthropic face mounting expectations, from governments, users, and civil society, to know who is using their tools and for what purposes.