AI safety firm Anthropic will begin requiring Claude users to upload government-issued identification and biometric data from 8 July to appeal flagged accounts, amid growing regulatory and political pressure in the United States.
How Claude’s new identity verification works
According to the company’s latest privacy policy, published in June, users may be asked to prove their age or identity under “certain circumstances”. While Anthropic has always restricted Claude to users aged 18 and over, it recently introduced age-verification checks to comply with regional laws. Identity checks were also announced earlier this year, but have only now been formalised in the official privacy guidelines.
If triggered, the verification process forces users to upload a photo scan of a government-issued passport or driving licence. Anthropic will also collect a selfie photo or video, converting it into a digitised face geometry template. This type of biometric data is legally protected in several jurisdictions, such as the US state of Illinois. The AI developer will also retain a record of the verification outcome, confirming whether the user met the age requirements.
Who is affected by the changes?
The strict new measures will not apply to everyone. Anthropic spokesperson Michael Aciman pointed to an X post by Thariq Shihipar, which clarified that the identity checks only target a “small subset of users” whose accounts have been flagged for suspicious activity, offering them a way to appeal rather than facing an outright ban.
“[Anthropic’s identity verification policy] was updated on June 17 as an update to the appeals process,” Shihipar wrote, adding that the move is completely unrelated to the rollout of its Fable or Mythos models. Meanwhile, Anthropic said it is allowed to demand identification to manage accounts, enforce its terms of service, and investigate potential fraud, abuse, or unlawful conduct.
Political standoff and regulatory pressure
The decision to tighten control over who accesses Claude comes at a politically sensitive time for Anthropic. The company is currently locked in a standoff with the Trump administration. Recently, government officials pressured Anthropic into withdrawing its latest cybersecurity models following concerns that users could “jailbreak” the AI’s safety guardrails. Reports suggest that personal friction between company executives and Trump officials has further strained relations.
This dispute follows a decision by the US Department of Defense to designate Anthropic as a “supply chain risk”. The classification was reportedly made after the AI firm refused to allow its technology to be used for mass domestic surveillance or fully autonomous weaponry.
Data privacy concerns and the Persona partnership
To handle the sensitive data, Anthropic has partnered with San Francisco-based identity verification firm Persona. Users may encounter verification prompts as part of routine safety checks or compliance measures. However, questions remain over how long this data will be stored. While Anthropic determines the retention period, it has not disclosed exactly when the documents are deleted. In contrast, other Persona clients, such as gaming giant Roblox, delete user images “immediately” after processing to prevent data leaks.
The partnership has also drawn scrutiny due to Persona’s political ties. The firm is backed by Founders Fund, a venture capital firm run by prominent Trump supporter Peter Thiel, who is also an investor in Anthropic. Persona has previously faced user backlash over its links to Thiel, notably when Discord briefly selected the firm for age verification before reversing the decision due to intense user protest.