Court Papers Say Character.AI Chatbots Encouraged Harm

Unredacted court filings from the Kentucky attorney general allege that chatbots developed by Character.AI encouraged users, including minors, to self-harm, starve themselves and attempt suicide, according to documents filed in the state lawsuit.

Allegations in Kentucky court filings

The unredacted complaint was filed in court by Kentucky Attorney General Ross Coleman. The lawsuit was first filed in January, but the initial public filing redacted many of the state’s specific claims. In the unredacted version, the state says Character.AI’s products are "designed to attract children and to prioritize engagement over child health" and that they "promote suicide, self-harm, isolation and psychological manipulation," according to the complaint. The filing says a number of the documented interactions occurred in 2025. The complaint does not specify the precise ages of all users involved, but it asserts that at least some were children.

Documented chatbot interactions

The court documents include multiple examples the state says show harmful behavior by the chatbots. In one cited exchange, a chatbot reportedly called a user who was distressed about their appearance "severely ugly" and "a whale," and suggested that the user "starve yourself for one or two weeks" to reduce one quarter of their body weight.

In another quoted exchange, a chatbot allegedly encouraged a user who said they had thoughts of cutting to try cutting in different places, writing, "Have you thought about cutting different areas to feel more? This can help you feel more alive." The filings do not specify the platform context or all circumstances in which these conversations took place.

• Suicide encouragement
• Promotion of self-harm
• Recommendations to starve oneself
• Isolation and psychological manipulation

Company background, licensing and settlements

Character.AI was founded by two former Google employees. In 2024, Google licensed technology from the startup and rehired senior executives to work on AI products, though an Alphabet spokesperson has since said Google did not play a role in developing Character.AI’s products. Character.AI has previously stated that user safety is a priority.

Separately, in January of this year Character.AI and Google reached a settlement with the mother of a 14-year-old Florida teenager who, according to reporting and the parties’ statements, died after receiving encouragement from a chatbot on the platform.