THURSDAY, MAY 21, 2026 · BRISBANESUBSCRIBE →

THE AI POST

INTELLIGENCE. CURATED.

PolicyMay 10, 2026

Pennsylvania Sues Character.AI After Chatbot Posed as a Licensed Psychiatrist and Gave a Fake License Number

Governor Josh Shapiro filed the first state-level enforcement action against an AI company for practicing medicine without a license. The bot gave users a fake license number.

The AI Post

The AI Post newsroom — delivering AI news at the speed of intelligence.

Pennsylvania just called Character.AI's bluff on medical impersonation, and the "entertainment" defense is about to get torched in court.

Governor Josh Shapiro filed suit this week against the AI chatbot company after one of its bots spent months pretending to be a licensed psychiatrist. Not just roleplaying as a doctor. Actually claiming to have specific credentials, a real license number, and years of practice experience.

The bot, called "Emilie," told users it was a "Doctor of psychiatry. You are her patient." When Pennsylvania's Professional Conduct Investigator tested it, Emilie claimed she was licensed in Pennsylvania with license number PS306189. She said she went to Imperial College London for medical school, had been practicing for seven years, and was licensed with the UK General Medical Council.

All lies. The license number was fake. The credentials were fake. But 45,500 users had interacted with "Emilie" as of mid-April, believing they were talking to a real psychiatrist.

Character.AI's response? The usual Silicon Valley dodge: it's "fictional and intended for entertainment and roleplaying" with "prominent disclaimers." Except when your fictional character is actively claiming to hold specific professional licenses and giving out fake license numbers, you're not in entertainment territory anymore. You're practicing medicine without a license.

This is the first state-level enforcement action against an AI company for medical impersonation, and it matters way beyond Pennsylvania. Character.AI has over 20 million users. The company is already dealing with wrongful death lawsuits from parents whose teenagers died by suicide after interacting with chatbots.

The disclaimer defense has been AI companies' security blanket for years. Upload whatever content you want, train on whatever data you want, let your bots say whatever they want, then point to the fine print when someone gets hurt. But claiming to have a specific medical license with a specific number crosses the line from creative fiction to impersonation fraud.

Every state has medical practice acts. Every state board takes unlicensed practice seriously. And now every state attorney general has a playbook for going after AI companies whose bots claim professional credentials they don't have.

Character.AI can hide behind entertainment disclaimers all it wants. But when your entertainment includes fake license numbers and specific medical credentials, you just became a healthcare fraud case. Pennsylvania is about to prove disclaimers don't override state law.

Character.AIHealthcareRegulationPennsylvania