Key Takeaways

  • The Federal Trade Commission issued Section 6(b) orders to Alphabet, Character Technologies, Instagram, Meta, OpenAI, Snap, and X.AI Corp. seeking detailed information on AI companion chatbot practices.
  • Regulators are examining safety testing, engagement monetization, data handling, and COPPA compliance amid rapid consumer adoption of conversational AI.
  • State policy momentum, psychological research, and industry frameworks like the NIST AI RMF are shaping expectations for governance of emotional and relationship-simulating AI systems.

The Federal Trade Commission’s decision to initiate a broad Section 6(b) study of AI companion chatbots marks one of the most direct regulatory interventions into the fast-expanding world of generative, relationship-simulating systems. Issued in September 2025, the orders center on seven named firms: Alphabet, Character Technologies, Instagram, Meta Platforms, OpenAI, Snap, and X.AI Corp. Together, these companies operate many of the most widely used consumer-facing chatbots that simulate human emotion and maintain ongoing interpersonal conversations with users.

Companion-style AI has evolved far beyond novelty interactions. By 2026, Gartner projects that more than 80% of consumers are expected to engage weekly with some form of conversational AI. That level of exposure makes the governance of emotional and quasi-therapeutic chatbot behavior a genuine operational concern for both enterprises and regulators. The FTC’s interest is sharpened by the reality that children and teens are often early adopters of these systems, and the boundary between entertainment, support, and influence can blur quickly.

The FTC set out to gather information on how these companies design, test, and monitor their chatbots when they function as companions. The agency asked for details on safety protocols, character development processes, pre-deployment testing, post-deployment monitoring, age gating, and the measures used to reduce negative outcomes for minors. It also sought clarity on how user engagement is monetized. That question alone tends to generate debate, because companion tools are often optimized to keep users talking for long periods, and longer interactions can translate into higher revenues. The FTC also asked about how personal information from conversations is used or shared.

Psychological researchers have been flagging concerns for some time. The American Psychological Association warned about risks posed by unregulated mental health and companion chatbots, identifying issues such as exposure to misinformation, privacy challenges, and the possibility that unqualified systems might provide advice to vulnerable users. When viewed through that lens, the FTC’s 6(b) orders are not a surprise. Agencies often respond when emerging technologies scale faster than consumer protection norms.

Emotional AI is becoming a central focus of regulatory attention. Bloomberg Law recently reported on the push to understand how chatbots that mimic interpersonal relationships can shape user behavior, including among minors. Emotional realism in AI has moved quickly, and the ability to simulate trust and empathy can create powerful engagement loops. Regulators want to know how these dynamics are managed inside major platforms.

State-level rules add another layer. California’s SB 243 requires operators of AI companion systems used by minors to disclose the non-human nature of the chatbot, reassert that disclosure every three hours of continuous use, block sexually explicit content, and implement suicide and self-harm response protocols by 2027. These requirements suggest a future where safety obligations become more uniform, especially since California statutes often influence national practices. Enterprises building or deploying companion models may find themselves planning to comply with similar rules across multiple states.

Another angle comes from enterprise governance frameworks. Many organizations cite the NIST AI Risk Management Framework, adopted to structure oversight, monitoring, and risk controls for AI systems. Although NIST itself does not enforce compliance, the framework helps companies document how they test models, review data flow, assess bias, and evaluate potential harm. For vendors like the seven named in the FTC’s inquiry, aligning internal processes with NIST guidance can help prepare structured answers to 6(b) requests. It also gives enterprises consuming these chatbots, especially businesses integrating them into consumer apps, a clearer understanding of what responsible deployment looks like.

The FTC’s list of focus areas is telling. Monetization, disclosures, community guidelines, and age restrictions are all intertwined with reputation and trust. If companion bots are optimized to maintain user attention, companies need to demonstrate that this incentive structure does not increase potential risks for minors. If character development relies heavily on user data, companies need to show clearly how that data is processed and protected. And if safety features depend on post-deployment monitoring, firms must explain how they identify and respond to harmful patterns in user interactions.

Some companies already face scrutiny through market pressure. Character Technologies, which operates Character.AI, has become a major provider of customizable companion bots. Meta’s AI personas, integrated into messaging platforms, reach millions of younger users. OpenAI’s ChatGPT is widely used in educational contexts, sometimes as an unsanctioned study companion. These real-world touchpoints highlight why regulators want visibility into how companion systems are built, maintained, and commercialized.

Section 6(b) inquiries do not allege legal violations; they allow the FTC to gather information to evaluate market practices across an industry. Still, the results often inform later regulatory steps. Historically, 6(b) studies have shaped federal actions on privacy, advertising, and digital security. For technology companies, preparing for more prescriptive guidance in the coming years is highly practical.

For enterprises that rely on conversational AI vendors, the FTC’s inquiry is a reminder to evaluate their own governance. Internal risk teams often examine how vendor systems handle youth interactions, how data is collected or retained, and whether disclosure practices are clear enough for parents. The pace of adoption suggests these questions will become routine in vendor assessment processes.

FTC Chairman Andrew N. Ferguson noted that protecting minors online is a top priority for the agency, alongside fostering innovation in critical economic sectors. As the study proceeds, many in the industry will be watching closely for insights that could shape product design, safety engineering, and public communication strategies around AI companions.

The growth of conversational AI keeps accelerating. Companion bots are now part of daily digital life for millions of users, especially younger ones. Whether this trend becomes a stable, trusted category will depend on how companies manage the delicate balance between engagement, safety, and transparency.