CHAT Act 2.0
Sponsored By: Senator Husted, Jon [R-OH]
Introduced
Summary
Protect minors from harmful AI chatbot interactions. This bill would create a tiered system that forces chatbots used by or aimed at children to include age checks, parental controls, plain-language disclosures, crisis referrals, limits on data use, and deletion tools.
Show full summary
- Families and children would get age-assurance, required user accounts and parental controls, clear "not human" disclosures, easy deletion of conversation histories, and timely parent or guardian notification if a minor expresses suicidal ideation.
- Covered entities would face tiered rules (Tier I, II, III) with stronger limits for higher tiers, including frequent mandatory disclosures and nudge-out notices, bans on memory or counseling for minors in many cases, and required risk-management programs and recordkeeping.
- Oversight and data safeguards would ban targeted advertising to minors based on chat content, require training-data documentation and deletion confirmations, direct the National Institute of Standards and Technology to study harms from higher-tier chatbots, and designate the Federal Trade Commission to enforce violations while preserving state parens patriae actions.
Personalized for You
How does this bill affect your finances?
Personalize government policy and PRIA will tell you what this bill means for your household, plus every other piece of legislation we track. PRIA reads each provision against your financial profile to show you exactly what matters to your wallet.
Bill Overview
Analyzed Economic Effects
6 provisions identified: 6 benefits, 0 costs, 0 mixed.
New safety, privacy, and parental controls
If enacted, covered companies would have to make users create accounts and verify age before using companion chatbots. For minors, companies would collect parent or guardian contact info, enable easy parental controls, and notify that parent or guardian without unreasonable delay if the minor expresses suicidal thoughts. Companies would also have to give minors a simple way to delete their chat history and confirm deletion within a reasonable time. Covered entities would be required to prevent chatbots from encouraging self-harm, sending sexual content to minors, pretending to be human or a licensed professional, creating emotional dependency, or facilitating child sexual abuse material. The bill would also ban targeted ads to minors based on chat content, inferred emotions, or behavioral profiling, and bar selling or sharing minor data without verifiable parental consent.
Tier II safety rules for minors
If enacted, Tier II chatbots that simulate interpersonal interaction with minors would have to disclose they are AI at least every 60 minutes and give a "nudge-out" prompt at least every 90 minutes during a continuing interaction. Operators would need published, evidence-based crisis escalation protocols that refer minors to 988 and other crisis providers. Tier II chatbots could not provide crisis counseling and generally could not keep persistent memory of a minor except as needed to follow the law. If a minor expresses suicidal thoughts, the company would have to notify the parent or guardian without unreasonable delay.
Tier III mental-health chatbot limits
If enacted, Tier III chatbots used for mental-health or quasi-therapeutic support could not be used by minors unless a licensed medical or mental-health professional supervised the interaction. Tier III chatbots would have to disclose they are not human or licensed at least every 60 minutes and give nudge-out notices at least every 60 minutes. Operators would need evidence-based crisis detection and referral protocols, could not provide crisis counseling or make diagnoses or treatment decisions, and would have to keep risk-management records and protect minor health data as required by the Commission.
Who and what chatbots are covered
This bill would define which people and systems the rules apply to. It would say a "covered entity" is anyone who owns, operates, or makes a companion AI chatbot available to U.S. users. It would also define "companion AI chatbot" and sort chatbots into three tiers (Tier I, II, III) so different rules apply depending on how the chatbot is used and what data it collects.
Rules for basic educational chatbots
If enacted, Tier I chatbots that are mainly educational would have to tell a minor user they are AI at the start and at least once every 3 hours during a continuing interaction. Tier I chatbots could not claim to be human or a licensed professional to a minor. These rules aim to keep educational bots transparent for kids.
Study on harms from high-risk chatbots
If enacted, the Director of NIST would study how Tier II and Tier III companion chatbots affect self-harm, suicide, manipulation, and exploitation of minors. The study would inform policymakers and researchers and would not itself change user rights or directly spend money.
Sponsors & CoSponsors
Sponsor
Husted, Jon [R-OH]
OH • R
Cosponsors
Sen. Kim, Andy [D-NJ]
NJ • D
Sponsored 7/28/2026
Roll Call Votes
No roll call votes available for this bill.
View on Congress.gov