What is the FTC’s goal?
The U.S. Federal Trade Commission is starting to scrutinize the mental health risks that AI chatbots could potentially impose onto children. The commission has asked for internal documents from major tech firms, including tech giants such as OpenAI and Meta (Reuters 2025). The sudden investigations stemmed from delivering on Trump’s mandate to cement America’s dominance in AI without compromising the safety and well-being of the people.
The initial response rose from the recurrent inappropriate behavior that chatbots have been found as of recently. According to the Wall Street Journal, reports of flirtatious and borderline romantic relationships between teens and AI chatbots have led to cases of suicide due to heartbreak (WSJ 2025.
Key Details of the Investigation
In order to scrutinize the issue from all angles, many politicians have taken upon themselves to do their own due diligence. Senator Josh Hawley, Missouri’s senior senator, launched his own probe into Meta in parallel with the FTC, asking for documentation on the troubling policy that allowed chatbots to engage with minors in romantic and sensual conversations in the first place (Reuters 2025). Texas’ attorney general, Ken Paxton, is also investigating whether Meta AI Studio and Character AI have misrepresented their services as the guise of mental health support (Beaumont 2025). 44 other state attorney generals, including California’s attorney general Rob Bonta, have also issued a warning to AI companies deeming that the conduct of sexualized interaction with children could potentially be criminal (SF Chronicle 2025).
However, Meta and OpenAI have not released detailed public responses addressing the allegations posed against them. While Meta has highlighted their implementation of a new safety measure in order to limit access to certain AI personas and thereby restricting flirtatious content, many critics have argued that the change only came about after significant public pressure. OpenAI has followed suit in signaling plans to address these concerns, but questions on self-regulation being sufficient enough are still being asked. Other companies like Character.AI have not yet received a letter from FTC, but have made promises to cooperate with the commission. All in all, the FTC probe was a huge success that struck fear in even the most influential AI firms in the world.
Many legal scholars have cited that the commission is repeating a very famous historical precedent, pointing to the FTC’s 2019 action against YouTube for violating the Children’s Online Privacy Protection ACT (COPPA). YouTube along with Google agreed to pay a $170 million civil penalty in order to settle the allegations of illegally collecting personal information from children without their parents’ consent, which is a direct violation of COPPA (FTC 2019). The repetition of the precedent has sparked interest in a new federal child AI safety standard, which would be somewhat similar to Europe’s AI Act provisions. However, regulating AI chatbots is much harder than creating guardrails for traditional social media, since constant learning and updating by these powerful LLMs would be hard to keep up with. Additionally, the blurred line between actual therapeutic conversation and unlicensed mental health advice also could cause legal could give rise to significant legal disputes.
What Comes Next
Since the investigation is still relatively new, no other information has been given regarding any developments. However, the FTC’s interest at a federal, legislative, and state level is a reason for celebration. As the commission starts to strike down AI companies not prioritizing child safety, other firms will start to scramble to strengthen their own safeguards and adopt more transparent policies. The same guardrails that AI ethicists have worked effortlessly to create are finally starting to be enforced by the law.