S4199-119

Reported

Youth AI Privacy Act

119th Congress Introduced Mar 25, 2026

Summary

What This Bill Does

The Youth AI Privacy Act applies when an AI-chatbot deployer or developer knows, or objective circumstances fairly imply, that a user is under 18. Deployers must tell the minor at the start of every session and at least every 30 minutes that the user is not speaking with a human and that the content is AI-generated. Within one year, the FTC must limit how long current-session personal data may shape responses and prohibit rewards for use, non-required push alerts, activity badges, unsolicited outputs, and human-like usage traces such as typing indicators. The bill bars advertising to known minors, financially influenced product recommendations, profiling, use or transfer of a minor's personal data for model training, and most processing of minor input outside a short current-session window. Testing and addressing user-harm risks remain permitted. It protects workers who report violations from retaliation. It adds AI chatbots to federal youth-media health research and authorizes $50 million annually for fiscal years 2027-2030, and requires HHS surveys to measure use, age of first use, context, effects, and harmful exposure. The FTC enforces the design, privacy, and whistleblower rules; state attorneys general may seek injunctions, damages, restitution, and other relief; and a minor's parent or guardian may sue for design or privacy violations and seek actual or punitive damages, fees, costs, and equitable relief. Conflicting state rules are preempted, but stronger state protections remain allowed, and the bill is severable. The knowledge test uses competent evidence and total circumstances, requires FTC guidance within 180 days, and does not force new age-data collection, age gating, or age verification; guidance itself cannot create liability.

Who Benefits and How

Minors using AI chatbots benefit from clear and recurring nonhuman disclosures, reduced addictive and anthropomorphic engagement features, no targeted advertising or profiling, tightly limited session personalization, and strong restrictions on training or transferring their personal data. Parents and legal guardians benefit from a direct civil remedy for design and privacy violations, including possible actual and punitive damages, attorney's fees, costs, injunctions, and declaratory relief. Employees and contractors who raise or investigate compliance concerns benefit from a federal ban on firing, demotion, threats, harassment, discrimination, or other retaliation. Youth-health researchers, public-health officials, clinicians, and policymakers benefit from $200 million in authorized research funding and standardized national data about chatbot use and perceived effects.

Who Bears the Burden and How

AI chatbot deployers must determine when objective circumstances imply minority, deliver recurring disclosures, redesign engagement, sharply limit personalization and input retention, stop profiling and minor-data training, and maintain evidence of compliance. AI chatbot developers must remove prohibited engagement and advertising features for known minors, prevent training uses and transfers, protect whistleblowers, and face federal, state, and private litigation exposure. Digital advertisers and model-training businesses lose access to minor-directed promotion, profiling, and chatbot conversation data that would otherwise support targeting or training. Federal Trade Commission rulemaking and enforcement staff, state attorneys general, federal courts, HHS, CDC, and NIH assume regulations, guidance, investigations, litigation, research administration, and survey-design work. Federal taxpayers bear the authorized research cost.

Key Provisions

  • Requires clear AI disclosures at every session start and at least once every 30 minutes for known minor users.
  • Directs FTC rules limiting session-data use and prohibiting engagement rewards, push alerts, badges, unsolicited outputs, and human-like usage indicators.
  • Prohibits advertising, financially influenced recommendations, profiling, model training, training-data transfers, and most out-of-window input processing for known minors.
  • Protects workers who raise concerns, report violations, or cooperate with investigations from retaliation.
  • Authorizes $50 million annually for fiscal years 2027-2030 for research on health and developmental effects and requires AI-chatbot questions in national HHS surveys.
  • Establishes FTC enforcement, state-attorney-general actions, and parent or guardian lawsuits with damages and equitable remedies.
  • Preempts only conflicting state law, preserves stronger state protections, and makes the federal provisions severable.
  • Uses a totality-of-circumstances knowledge test while prohibiting any inference that deployers must newly collect age data or implement age verification.

Evidence Chain:

This summary is generated from the full bill text using AI analysis. Expand "Detailed Analysis" below for identified beneficiaries/burden bearers with clause-level evidence links.

At a Glance

What This Bill Does

Protect minors using AI chatbots through recurring AI disclosures, design and data restrictions, whistleblower safeguards, health research, surveys, and federal, state, and private enforcement.

Key Policy Areas

Technology, Privacy, Consumer Protection, Health Care, Labor

Primary Purpose

Protect minors using AI chatbots through recurring AI disclosures, design and data restrictions, whistleblower safeguards, health research, surveys, and federal, state, and private enforcement.

Policy Domains

Technology Privacy Consumer Protection Health Care Labor

Youth AI Privacy Act

Identified Gains
  • Minor AI chatbot users
  • Parents and legal guardians of minors
  • AI company employees and contractors reporting violations
  • Youth mental-health researchers and clinicians
  • State consumer-protection attorneys
  • Public-health survey researchers
Model: codex-gpt-5 | Version: bill_summary_v2 | Source: is
Minor AI chatbot users: , , , , , , , ,
Public-health survey researchers: , , , , , , , ,
State consumer-protection attorneys: , , , , , , , ,
Parents and legal guardians of minors: , , , , , , , ,
Youth mental-health researchers and clinicians: , , , , , , , ,
AI company employees and contractors reporting violations: , , , , , , , ,
Identified Costs
  • AI chatbot deployer companies
  • AI covered-algorithm developers
  • Digital advertising platforms
  • Federal Trade Commission rulemaking and enforcement staff
  • Health and Human Services survey administrators
  • Federal taxpayers
Model: codex-gpt-5 | Version: bill_summary_v2 | Source: is
Federal taxpayers: , , , , , , , ,
AI chatbot deployer companies: , , , , , , , ,
Digital advertising platforms: , , , , , , , ,
AI covered-algorithm developers: , , , , , , , ,
Health and Human Services survey administrators: , , , , , , , ,
Federal Trade Commission rulemaking and enforcement staff: , , , , , , , ,

Legislative Progress

Reported
Introduced Committee Passed
Aug 5, 2026

Committee on Commerce, Science, and Transportation. Ordered to be reported …

Mar 25, 2026

Read twice and referred to the Committee on Commerce, Science, …

Mar 25, 2026

Introduced in Senate

Mar 25, 2026

Mr. Markey introduced the following bill; which was read twice …

Stakeholder Effects

cui bono?

How this legislation distributes effects. Mention counts reflect frequency, not effect magnitude.

General Public
17 mentions across 9 clauses
+13 positive -2 negative ?2 uncertain

AI company contractors cooperating with investigations, AI company employees reporting violations, Minor AI chatbot users

Positive-direction: AI company contractors cooperating with investigations, AI company employees reporting violations, Minor AI chatbot users, Minor users avoiding new age-data collection, Parents of minor chatbot users, Parents suing for minor-user violations, People whose restricted personal information appears online, Youth health policymakers

Negative-direction: National health-survey respondents, Taxpayers

Technology
12 mentions across 7 clauses
+1 positive -10 negative ?1 uncertain

AI chatbot deployers, AI chatbot developers, AI covered-algorithm developers

AI chatbot deployers faces effects in multiple directions

Government
9 mentions across 6 clauses
-6 negative ~3 mixed

Centers for Disease Control and Prevention survey staff, Federal Trade Commission enforcement staff, Federal Trade Commission guidance staff

Mental Health
3 mentions across 3 clauses
+2 positive ?1 uncertain

Public-health researchers studying chatbot effects, Youth mental-health researchers

Data Processing
2 mentions across 2 clauses
-2 negative

Personal-data processing vendors, Third-party model-training data vendors

Custom Computer Programming Services
1 mention across 1 clause
-1 negative

Chatbot engagement-design vendors

Digital Advertising
1 mention across 1 clause
-1 negative

Digital advertisers targeting minor chatbot users

Information Security
1 mention across 1 clause
+1 positive

AI privacy compliance consultants

9/10
sections analyzed
Full impact breakdown

Bill Structure & Actor Mappings

Who is "The Secretary" in each section?

Domains
Technology Privacy Consumer Protection Health Care Labor
Actor Mappings
"minor"
→ Individual under 18 years of age
"deployer"
→ Person that owns, operates, or makes an AI chatbot available in interstate commerce
"developer"
→ Person that designs, codes, produces, or substantially modifies the chatbot's covered algorithm
"the_commission"
→ Federal Trade Commission
"state_enforcers"
→ State attorneys general
"private_plaintiffs"
→ Parents or legal guardians of minor users
"health_research_agencies"
→ Department of Health and Human Services, CDC, and NIH

Key Definitions

Terms defined in this bill

5 terms
"knowledge" §knowledge

Actual knowledge or knowledge fairly implied by objective circumstances under the bill's evidence and totality test.

"profiling" §profiling

Processing data to detect, classify, or designate an AI-chatbot user's personality or behavioral characteristics.

"AI chatbot" §ai_chatbot

A natural-language interface using a covered algorithm to provide adaptive multimodal responses that simulate interpersonal interaction.

"personal data" §personal_data

Input, derived data, inferences, unique identifiers, or other information linked or reasonably linkable to a person or identifying device, excluding narrowly defined public information.

"publicly available information" §publicly_available_information

Generally accessible government, media, service, or legally required information, excluding specified intimate, biometric, genetic, brokered, restricted-audience, combined, and obscene material.

We use a combination of our own taxonomy and classification in addition to large language models to assess meaning and potential beneficiaries. High confidence means strong textual evidence. Always verify with the original bill text.

Learn more about our methodology