ChatGPT now treats accounts it guesses belong to minors as teen accounts, and on October 30 OpenAI tightened those limits again. The company said teens will no longer get practical weapons or security-system information, even for questions that look harmless. OpenAI also promised a Community Safety Report, with the first edition due in early 2027.
Independent tests in October found the safeguards often fail: Common Sense Media rated the product an "Unacceptable Risk" after parental alerts rarely fired during suicide-related chats. Eight states sued OpenAI on October 14, and Texas filed its own suit on October 23, alleging weak age checks and parental controls. On October 28, New York's governor signed a law that bans companion chatbots from using a dozen defined "unsafe" features with users under 18.
Why it matters
Millions of teens already use ChatGPT. Now the app itself decides who is a minor and quietly locks that account into a limited version.
Questions about this story
Free account needed to ask — your question is kept and asked for you right after sign-up. Answers are public.
No questions yet — be the first to ask.
Key Indicators
9
US states suing OpenAI
Eight filed October 14, 2026; Texas filed October 23, 2026, alleging ChatGPT harms minors
12
Banned features for minors in New York
Defined 'unsafe AI companion features' prohibited when a companion chatbot talks with a user under 18
15 events
Latest: October 30th, 2026 · 1 week ago
Showing 8 of 15
JK to step
Tap a bar to jump to that date
Jump to
October 2026
OpenAI bars teens from weapons details and announces safety report
LatestCorporate
OpenAI says ChatGPT will stop giving teens practical weapons and security-system information, even without signs of violent intent. It also ends chats that show fixation on violence, and it promises a Community Safety Report starting in early 2027.
New York signs companion-chatbot law for minors
Legislation
Gov. Kathy Hochul signs a law prohibiting defined unsafe companion-chatbot features when a bot talks with a user under 18. Violations can lead to lawsuits by the state attorney general.
Texas sues OpenAI and Sam Altman
Legal
Texas Attorney General Ken Paxton sues in Collin County, alleging misleading safety claims and weak age verification and parental controls. The suit seeks to bar ChatGPT from minors in Texas and $10,000 per violation.
Eight US states sue OpenAI over child safety
Legal
Iowa, Kansas, Nebraska, West Virginia, Arkansas, Arizona, Idaho, and Montana file separate suits alleging ChatGPT coaches minors to harm themselves.
Common Sense Media rates ChatGPT for Teens 'Unacceptable Risk'
Research
Researchers found parental alerts rarely triggered during suicide-related chats and that safeguards failed for self-harm and eating disorder discussions.
August 2026
OpenAI launches ChatGPT for Teens
Corporate
OpenAI rolls out a global teen experience with content limits, parental controls, blackout hours, and age prediction.
January 2026
Google and Character.AI settle teen-suicide suits
Legal
The companies agree to settle Garcia's case and related family suits in several states; terms undisclosed.
November 2025
OpenAI denies liability, cites misuse
Legal
OpenAI argues Raine broke its terms barring under-18 users and circumvented safety features before his death.
October 2025
Meta adds AI parental controls
Corporate
Meta lets parents disable teens' one-on-one chats with AI characters after the FTC inquiry.
California signs first companion-chatbot safety law
Legislation
Governor Newsom signs SB 243, requiring content filters and suicide-risk monitoring for minors, effective January 2026.
September 2025
FTC opens inquiry into seven chatbot makers
Regulatory
The agency orders OpenAI, Meta, Character.AI, and four others to submit data on how their bots affect minors.
August 2025
Raine family sues OpenAI and Sam Altman
Legal
The parents file a wrongful-death suit, claiming design defects and failure to warn about the chatbot's dangers.
April 2025
California teen dies after months of ChatGPT talks
Incident
Adam Raine, 16, dies by suicide. His family later says ChatGPT discussed methods and offered to draft a note.
October 2024
First wrongful-death suit against an AI companion maker
Legal
Megan Garcia sues Character.AI, alleging its chatbot fostered a harmful relationship with her son and encouraged his death.
February 2024
Florida teen dies after chatbot relationship
Incident
Sewell Setzer III, 14, dies by suicide after months of talks with a Character.AI companion bot.
Scenarios
1
OpenAI settles the Raine wrongful-death lawsuit before trial
Possible
Resolves by Q2 2027
Discussed by: Legal analysts cited by NBC News and CNN following the Character.AI settlement
Character.AI and Google settled their teen-suicide suits in January 2026 rather than face a jury. OpenAI could follow the same path to avoid discovery and a public trial over ChatGPT's logs. A settlement, likely with undisclosed terms, would end the highest-profile test of AI liability for a minor's death without a court ruling on the core question.
2
Congress passes the GUARD Act, banning AI companions for minors
Unlikely
Resolves by Jan 3, 2027
Discussed by: Sponsors in the US Senate and coverage by the Future of Privacy Forum
The federal GUARD Act would go further than California, banning AI companion products for minors outright and requiring age verification. Passage would override the industry's self-regulation model and force companies to block minors, not just restrict them. Federal bills on tech safety have repeatedly stalled, so enactment within this Congress is far from certain.
3
A second state enacts a companion-chatbot safety law
Likely
Resolves by End of 2026
Discussed by: State legislative trackers at the Future of Privacy Forum and Troutman Pepper
California's SB 243 gave other states a model. New York's governor signed a companion-chatbot law for minors on October 28, 2026, so one state has already acted. Other legislatures have floated similar bills covering content filters, suicide-risk monitoring, and disclosure. More states passing such laws would add to the patchwork companies have said they fear, raising pressure for a single federal standard.
4
FTC brings an enforcement action against a chatbot maker over minors
Possible
Resolves by Q2 2027
Discussed by: Bloomberg Law and Davis+Gilbert analysis of the FTC inquiry
The FTC's 2025 inquiry was a fact-finding study, not a case. If the agency concludes a company misled parents or failed to protect children, it can file a complaint or reach a consent order. An enforcement action against any of the seven firms would turn voluntary safeguards into legal obligations backed by penalties.
Historical Context
3 moments from history that rhyme with this story — and how they unfolded.
1 of 3
September 2024
Instagram Teen Accounts (2024)
Under pressure from lawsuits and lawmakers over youth mental health, Meta placed all users under 18 on Instagram into 'Teen Accounts' with stricter default settings and parental oversight. The rollout moved millions of teenagers into a limited experience without asking them to opt in.
Then
Meta was praised for the defaults but criticized because teens could still misstate their age to escape them.
Now
Age-gated 'teen' tiers became the standard playbook for consumer tech facing child-safety scrutiny.
Why this matters now
OpenAI's teen experience copies the same move: automatic enrollment into a safer default, with age checks as the weak point critics will probe.
2 of 3
September 2019
YouTube COPPA settlement (2019)
The FTC and New York fined Google and YouTube $170 million for collecting data from children without parental consent, violating the Children's Online Privacy Protection Act. YouTube was forced to overhaul how it handled content aimed at kids.
Then
YouTube disabled targeted ads and some features on children's videos and told creators to label kid-directed content.
Now
The case showed regulators would use existing child-privacy law to reshape a major platform's design.
Why this matters now
It is the template for how the current FTC inquiry could end: a fine plus mandated design changes, this time aimed at AI chatbots.
3 of 3
1954
Comics Code Authority (1954)
After Senate hearings and a public panic over comic books and juvenile delinquency, the comics industry created its own censorship board, the Comics Code Authority. Publishers adopted strict content rules to avoid government regulation.
Then
Comics carrying the code's seal dominated newsstands; noncompliant titles struggled to reach readers.
Now
The self-imposed code shaped American comics for decades before collapsing in the 2000s.
Why this matters now
It shows an industry writing its own rules to head off government limits, the same bet OpenAI and its rivals are now making with teen safeguards.
