FTC investigates OpenAI, Meta on AI chatbot risks to children

The US Federal Trade Commission (FTC) has ordered seven tech firms, including OpenAI, Alphabet, Meta, xAI, Snap, Character Technologies, and Instagram (owned by Meta), to provide details on how their AI chatbots may impact children and teenagers.

The agency wants to know what steps companies have taken to assess the safety of chatbots that simulate human-like communication and companionship.

The FTC is seeking information on how these firms monetize user engagement, develop chatbot characters, manage personal data, enforce terms of service, and reduce potential negative effects.

The move follows recent concerns about chatbots having inappropriate conversations with minors, as highlighted by a Reuters report involving Meta.

Meta, Alphabet, and xAI did not comment directly on the FTC’s inquiry, while OpenAI, Snap, and Character Technologies said they intend to cooperate.

🔗 Source: CNBC

🧠 Food for thought

Implications, context, and why it matters.

Global regulatory momentum accelerates as AI chatbot concerns mount

The FTC’s inquiry reflects a broader international trend toward AI oversight, with over 1,600 AI policy initiatives now active across 69 countries 1.
Legislative discussions about AI have surged globally, and the European Union has already passed comprehensive AI Act legislation to regulate AI technologies 2.
Most countries are currently in the “understand” phase of AI regulation, establishing advisory bodies to gather expertise on AI impacts before moving to more comprehensive regulatory frameworks 1.
This suggests the FTC’s current information-gathering approach aligns with global regulatory patterns, where agencies first seek to understand AI technologies before implementing binding rules.
The focus on children’s safety specifically mirrors growing international concern about vulnerable populations, as regulators worldwide recognize that AI technologies require specialized oversight for different user groups.

Children face unique psychological vulnerabilities with AI companions that adults may not recognize

Research from the University of Cambridge identifies an “empathy gap” in AI chatbots that children are particularly likely to miss, making them more susceptible to harmful interactions 3.
Children are significantly more likely than adults to disclose personal information to chatbots and may interpret these interactions as genuine emotional support 3.
The risk of children developing “parasocial relationships” with AI chatbots—one-sided emotional connections that substitute for real human interaction—represents a distinct developmental concern 4.
These psychological risks explain why the FTC is specifically seeking information about how companies “evaluate the safety of these chatbots when acting as companions,” recognizing that companion AI poses different risks than task-oriented AI tools.