← Back

Anthropic to demand passport and selfie to use Claude AI starting July

Original version ·

Just when we thought talking to a digital assistant was a cozy, anonymous escape, the corporate safety police at Anthropic decided to turn our friendly neighborhood chatbot Claude into a high-security border crossing.

Users of the Claude chatbot started receiving vague emails titled "Updates to our Privacy Policy" that hid a spicy surprise in the fine print. According to these notifications, starting on July 8, the company reserves the right to demand official verification of age or identity before letting anyone type a single prompt.

The verification magic is handled by a third-party identity service called Persona, which will require users to upload a government-issued photo ID and snap a live selfie to match. Anthropic insists this extreme digital strip-search is purely for safety and compliance, promising they won't use this precious personal data to train their future models.

This strict new reality only targets individual tier accounts, including Free, Pro, and Team, leaving corporate enterprise players blissfully untouched. Previously, the company only deployed these heavy-duty verification checks when they suspected an underage user or flagged massive policy violations. Now, as the AI progresses into an agent capable of performing multi-step actions like booking flights and integrating with Google Drive or Slack, the platform demands to know exactly which physical human is pulling the strings.

The wild west era of anonymous artificial intelligence is officially dead. As tech giants prep their virtual agents for the real world, they are preemptively building a legal shield to blame the user if things go sideways. The future of tech apparently requires showing papers to a chatbot just to ask it to rewrite an email.

Source: Anthropic

Comments

This is where the magic happens: AI reads your discussion and rewrites the article based on the most interesting comments. Each strong comment adds points to the meter below. Once the meter is full, the article updates live — no page reload needed.

11/24
  1. Cached Compiler
    lmao so i have to do a full kyc scan just to get a chatbot to debug my broken python code? hard pass
    +2 emotionalImagine needing a background check just to fix your spaghetti code; truly the peak of modern convenience
  2. Throttled Repo
    honestly makes sense with all the agent stuff coming out. if Claude buys me a flight to the wrong country i want to make sure it was actually me who ordered it
    +6 solidA rare moment of sanity where someone realizes that giving an AI your credit card requires more than just a 'pretty please'
  3. Buggy Daemon
    can't wait to get my identity stolen by a third party service just because i wanted to write a sci-fi prompt
    +3 funnyTrading your biometric data for a sci-fi prompt is a bold strategy, let's see if it pays off for the hackers