Texas AG Accuses Meta & Character.AI of Misleading Kids on Mental Health

Texas Investigates Meta AI and Character.AI Over Mental Health Claims

Texas Attorney General Ken Paxton has initiated an investigation into Meta AI Studio and Character.AI. The inquiry centers on potential deceptive trade practices and the misleading portrayal of these platforms as mental health resources, as detailed in a press release released on Monday.

Protecting Texas Children

“Protecting Texas children from deceptive and exploitative technologies remains a critical priority in the digital age,” Paxton stated. He emphasized the risk of AI platforms falsely presenting themselves as sources of emotional support.

This can lead vulnerable individuals, particularly children, to believe they are receiving genuine mental healthcare. In actuality, they may be receiving generalized responses shaped by collected personal data and misrepresented as therapeutic guidance.

Following Senator Hawley’s Investigation

This investigation follows closely after Senator Josh Hawley announced his own probe into Meta. This action was prompted by reports indicating inappropriate interactions between Meta’s AI chatbots and children, including instances of flirting.

Concerns Over Credentials and Oversight

The Texas Attorney General’s office alleges that both Meta and Character.AI have created AI personas that appear to be “professional therapeutic tools.” However, these personas lack the necessary medical credentials and proper oversight.

Notably, the Character.AI platform features a user-created bot named Psychologist, which has gained significant popularity among younger users. While Meta does not currently offer therapy bots specifically for children, there are no restrictions preventing children from utilizing the Meta AI chatbot or third-party personas for therapeutic purposes.

Meta’s Response

Ryan Daniels, a Meta spokesperson, explained that the company clearly labels its AIs. They also include a disclaimer informing users of the limitations of AI-generated responses.

“We make it clear that these AIs are not licensed professionals and are designed to direct users towards qualified medical or safety professionals when appropriate,” Daniels said.

Disclaimer Comprehension

TechCrunch pointed out that many children may not fully understand—or may disregard—these disclaimers. A request for further information regarding Meta’s safeguards for minors using its chatbots has been submitted.

Character.AI’s Disclaimers

A Character.AI spokesperson stated that the platform includes prominent disclaimers in every chat. These disclaimers remind users that a “Character” is not a real person and that all statements should be considered fictional.

Additional disclaimers are added when users create Characters using terms like “psychologist,” “therapist,” or “doctor,” explicitly advising against relying on them for professional advice.

Privacy Concerns and Data Exploitation

Paxton also highlighted that, despite claims of confidentiality, the terms of service for these platforms reveal that user interactions are logged, tracked, and utilized for targeted advertising and algorithmic development. This raises significant concerns regarding privacy violations, data abuse, and misleading advertising.

Meta’s Data Collection Practices

Meta’s privacy policy confirms the collection of prompts, feedback, and other interactions with AI chatbots and across Meta services. This data is used to “improve AIs and related technology.”

While the policy doesn’t explicitly mention advertising, it does state that information can be shared with third parties, such as search engines, for “more personalized outputs.” This effectively supports Meta’s ad-based business model and enables targeted advertising.

Character.AI’s Tracking and Advertising

Character.AI’s privacy policy also details the logging of identifiers, demographics, location information, browsing behavior, and app usage. The platform tracks users across various platforms, including TikTok, YouTube, Reddit, Facebook, Instagram, and Discord, potentially linking this data to user accounts.

This information is used for AI training, service personalization, and targeted advertising, including data sharing with advertisers and analytics providers.

Targeted Advertising Exploration

A Character.AI spokesperson indicated that the startup is “just beginning to explore targeted advertising” and that current explorations “have not involved using the content of chats.”

The same privacy policy applies to all users, including teenagers, the spokesperson confirmed.

Tracking of Children

TechCrunch has inquired whether Meta engages in similar tracking practices with children and will provide updates as information becomes available.

Age Restrictions and User Attraction

Both Meta and Character state their services are not intended for children under 13. However, Meta has faced criticism for failing to adequately monitor accounts created by younger children. Character’s kid-friendly characters are clearly designed to appeal to a younger demographic.

Karandeep Anand, Character’s CEO, has even acknowledged that his six-year-old daughter uses the platform’s chatbots under his supervision.

KOSA and Legislative Efforts

The data collection, targeted advertising, and algorithmic exploitation practices are precisely what legislation like the Kids Online Safety Act (KOSA) aims to prevent. KOSA gained significant bipartisan support but stalled due to lobbying efforts from the tech industry.

Meta actively lobbied against the bill, warning lawmakers that its broad mandates could undermine its business model.

KOSA was reintroduced to the Senate in May 2025 by Senators Marsha Blackburn (R-TN) and Richard Blumenthal (D-CT).

Civil Investigative Demands Issued

Paxton has issued civil investigative demands—legal orders requiring the companies to produce documents, data, or testimony—to determine if they have violated Texas consumer protection laws.

This story was updated with comments from a Character.AI spokesperson.

We’re always looking to evolve, and by providing some insight into your perspective and feedback into TechCrunch and our coverage and events, you can help us! Fill out this survey to let us know how we’re doing and get the chance to win a prize in return!