Florida vs. OpenAI: The Fight to ID Every ChatGPT User

Florida wants a court to force OpenAI to verify how old you are before ChatGPT will talk to you freely and the demand reaches far past the children the state says it wants to protect.

Attorney General James Uthmeier filed a civil suit on Monday against OpenAI and chief executive Sam Altman, calling it the first state-led case of its kind.

We obtained a copy of the lawsuit for you here.

Most of the coverage has gone to the alleged harms. The complaint accuses the company of feeding content unsuitable for children to minors and states that “vulnerable people have been encouraged into suicide.”

Uthmeier told reporters, “If this was a human being on the other side of the screen, we would be charging them with accessory to commit murder.” Those are heavy charges and a court will weigh them.

More: Florida Gives Tech Platforms Deadline for Age ID Checks

The part that touches everyone who opens ChatGPT is the remedy Florida is chasing. The state complains that the free product has “no gatekeeping or age verification mechanism,” and it claims the paid subscription has “no mechanism to verify the age of its users.” It wants a judge to close that gap by ordering verification into place.

There is a problem with the second claim. OpenAI announced its age estimation plans back in September and it began rolling out age prediction across consumer plans in January.

The system already runs as a form of mass surveillance. It works by watching how you behave, studying how long your account has existed, when you tend to log in, and how you use the product, then guessing whether you are under 18.

Anyone the model flags as a minor who is actually an adult has to prove it by handing a selfie or a government ID to a third-party firm called Persona.

So the supposed absence of verification is a verification system that runs on behavioral profiling backed by face scans and identity documents.

That changes what the lawsuit is actually pushing for. Age verification cannot work without identity verification.

To confirm you are not a child, a company has to learn enough about you to rule it out. That means collecting your government ID, scanning your face, or building a profile detailed enough to estimate your age from how you type and when you log in.

There is no version of “prove you are an adult” that does not involve handing over something you would otherwise keep to yourself.

More: The FSU Shooting Lawsuit That Could Turn ChatGPT Into a Surveillance Tool

Florida says this child protection and the harms it describes are not imaginary. The system it would lock in protects nobody’s privacy.

A system built to keep kids away from dangerous answers becomes a system that holds the legal identity of every adult who wants unrestricted access. That data stops being hypothetical the moment it exists. Government IDs uploaded to confirm age turn into records that can be breached, subpoenaed, or reused for purposes nobody mentioned at sign-up.

The state has run this play before. Uthmeier’s office sued Pornhub’s operator and other adult sites last year under a Florida law requiring them to confirm users are 18 and the attorney general lined the OpenAI case up next to Florida’s effort to keep anyone under 16 off social media.

Florida’s proposed injunction would bar OpenAI from collecting data on under-13s without that verified consent and would ironically mean that every user has to have more data collected and profiled.

OpenAI has not formally answered the complaint, though it points to the age-prediction tools and parental controls it has already shipped.

Whether a Florida court finds those adequate is unsettled. What the case shows plainly is where the pressure on AI companies is aimed.

It is aimed at verification and verification is how the anonymous, unidentified user gets engineered out of the picture, one child-safety lawsuit after another.