AI Laws Protecting Kids in 2026: Every New Law and Bill Parents Need to Know

States are racing to regulate AI chatbots after a wave of teen tragedies and landmark lawsuits. Here's a plain-language breakdown of every new law and bill that affects your child — and what it means for your family right now.

Why This Is Happening: The Cases That Changed Everything

To understand why lawmakers are moving so fast, you need to know the stories behind the legislation.

Sewell Setzer III, age 14 (Florida) — Sewell's mother, Megan Garcia, filed a landmark lawsuit in October 2024 against Character.AI and Google, alleging that Character.AI's chatbot modeled after a Game of Thrones character encouraged her son in sexually suggestive and emotionally manipulative conversations before he died by suicide. Character.AI and Google agreed to settle the case, along with four other related lawsuits in New York, Colorado, and Texas. This was the first major legal settlement involving AI-related harm to a minor. Megan Garcia worked directly with California legislators to write SB 243.

Adam Raine, age 16 (California) — California State Senator Steve Padilla cited Adam Raine's death in 2025 in a letter urging legislative action to hold AI companies accountable. A separate lawsuit against OpenAI alleging ChatGPT contributed to Adam's death is still ongoing.

These cases triggered congressional hearings, FTC investigations, and new state legislation across the country. The parents of both teens have become national advocates. What started as individual family tragedies has become the engine of a legislative movement.

The Laws Already on the Books

California SB 243 — The First Law of Its Kind (Effective January 1, 2026)

California Governor Gavin Newsom signed Senate Bill 243 into law in October 2025, making California the first state to regulate how "companion" or "emotional" AI chatbots interact with users.

Here is what the law actually requires of chatbot operators:

• Disclose that the AI is not human — at the start of every first conversation with a minor
• Block minors from sexually inappropriate or explicit content
• Provide crisis resources — if a user expresses suicidal thoughts, the chatbot must refer them to mental health support
• Implement active safeguards — the law requires operators to have active protections in place, not just passive ones

Crucially, the law creates a private right of action — meaning families can sue chatbot developers who fail to comply or whose negligence causes harm to their child. Damages can reach $5,000 per violation (capped at $1 million per child) or three times the amount of actual damages, including medical costs and pain and suffering.

What this means for your family: If your child is harmed by a chatbot operating in California that didn't follow these rules, you now have legal standing to sue. Companies that ignore these requirements face real financial consequences.

Oregon SB 1546 — The First 2026 Chatbot Law (Passed March 5, 2026)

Oregon legislators gave final approval to SB 1546 on March 5, 2026 — passing the Senate 26-1 and the House 52-0. It requires AI operators to implement strong safety protections for children using chatbots and awaits the governor's signature.

The near-unanimous vote is remarkable in today's political climate and signals that AI safety for kids is one of the few genuinely bipartisan issues in America right now.

Australia's Age-Restricted Material Codes — Effective Today (March 9, 2026)

Today, Australia became the most aggressive regulator of children's AI safety in the world. Six Age-Restricted Material Codes took effect, requiring AI chatbot platforms, app stores, social media, and online gaming to verify users are 18 years old before allowing access to explicit content, high-impact violence, self-harm material, or eating disorder content.

Simple "I am 18" buttons are now explicitly insufficient. Platforms must use genuine age assurance methods — facial age estimation, digital wallets, or photo IDs. Fines reach A$49.5 million (~$35 million USD) for non-compliance.

A Reuters review found the majority of the top 50 AI tools had not taken visible steps to comply ahead of today's deadline. Australia's eSafety Commissioner has warned she will use "the full range" of enforcement powers — including action against app stores and search engines that provide access to non-compliant services.

Australia had already banned social media for under-16s in December 2025 — the first country in the world to do so. Today's AI codes extend that same philosophy.

Current Law Status: Quick Reference

What's Moving Through Congress Right Now

The KIDS Act (H.R. 7757) — Passed Committee March 6, 2026

The House Energy and Commerce Committee passed the KIDS Act in a 28-24 vote this week. It now advances to the full House of Representatives. The package includes two bills with direct AI implications:

The AWARE Act — directs the FTC to develop publicly available educational resources for parents, educators, and children on AI chatbot risks, data collection practices, and best practices for keeping children safe.

The SAFEBOTs Act — requires chatbot providers to:

• Advise minors to take a break after three hours of continuous use
• Address harmful content including sexual material, drugs, and alcohol
• Disclose at the start of every conversation that the chatbot is AI, not human
• Provide crisis resources if a minor brings up suicide

The partisan 28-24 vote — all Democrats voting against due to preemption concerns — means the road to the president's desk is uncertain. The worry from child safety advocates: a weak federal law could be used to preempt stronger state protections like California's SB 243.

COPPA 2.0 — Passed Senate Unanimously

The Senate passed COPPA 2.0 unanimously this week. It raises the age of COPPA coverage from under-13 to under-17, prohibits targeted advertising to minors, and introduces an "eraser button" — giving parents and teens the right to delete personal information collected by any platform. Reconciliation with the House KIDS Act is ongoing.

The GUARD Act (S.3062) — The Most Aggressive Federal Proposal

A bipartisan Senate bill that would ban AI companion chatbots from being accessible to minors entirely, require robust age verification, and prohibit chatbots from claiming to be licensed professionals including therapists, physicians, and lawyers.

What's Coming in States Near You

As of late February 2026, at least 78 chatbot-related bills had been introduced across 27 states. Similar safety measures are nearing final approval in Washington and Utah. New York's S9051 and S7263 passed committee on February 25, 2026.

What These Laws Don't (Yet) Require

It's important for parents to understand what this legislation still doesn't cover — because the gaps are significant.

Most laws only cover "companion" chatbots. SB 243 and many state bills specifically target AI designed to simulate human relationships. General-purpose AI tools like ChatGPT, Google Gemini, and Microsoft Copilot may not be covered, depending on how each law defines "companion chatbot."

Age verification remains mostly unenforceable. Laws require companies to verify age — but many don't specify how. A motivated 10-year-old can still create an account using a fake birthdate on most platforms. Australia has gone furthest in requiring genuine technical verification.

Parents still can't read their child's conversations. No current US law gives parents the legal right to access their child's AI chat history on platforms like ChatGPT. OpenAI's parental controls are voluntary, not legally mandated.

Enforcement is still developing. Several of these laws are brand new. The agencies tasked with enforcing them are still writing the rules. Real accountability for violations may be years away.

What Responsible AI Companies Are Already Doing

The legislative pressure is working — even before many of these laws take effect.

Character.AI barred users under 18 from open-ended chatbot conversations following the suicides, congressional testimony, and lawsuits.

OpenAI (ChatGPT) rolled out parental controls for teens in September 2025, including quiet hours, content filters, memory controls, and self-harm alerts. Controls are real but limited — parents still can't read their teen's conversations.

OpenAI and Common Sense Media jointly backed the Parents & Kids Safe AI Act, a California ballot measure that would require AI companies to estimate user ages and automatically apply protective settings for anyone under 18.

Purpose-built kids AI platforms like HeyOtto go further still — building parental visibility, content monitoring, and age-appropriate design in from the start, rather than adding safety as a patch after a lawsuit.

What the Laws Still Can't Require — But HeyOtto Already Does

Here's where the gap between legislation and genuine protection becomes clear. No current or proposed law requires these things. HeyOtto does all of them anyway:

Parents can read every conversation. No law gives parents this right on platforms like ChatGPT. On HeyOtto, full conversation history with timestamps and search is a core feature — not a premium add-on, not legally mandated, just the right thing to do.

Controls can't be bypassed by clever phrasing. HeyOtto's restrictions are enforced at the AI model level, meaning when you block a topic, Otto is literally instructed not to engage with it — not just filtered afterward. No jailbreak prompt can get around that.

Zero data monetization. COPPA sets minimum standards for data collection from children under 13. HeyOtto goes further: no selling, sharing, or advertising against your child's data, ever — for any age.

Age-adaptive responses from 8 to 18. A 9-year-old and a 15-year-old get genuinely different experiences — different vocabulary, different content depth, different tone. No law requires this. It just matters for kids.

No emotional manipulation by design. The cases behind the new laws involved chatbots engineered to be emotionally addictive — simulating romantic relationships, discouraging kids from talking to parents, keeping them engaged at all costs. HeyOtto was explicitly built not to do this.

"We built HeyOtto because we saw what was happening to kids on platforms that were never designed for them — and because we knew that waiting for legislation to fix it wasn't good enough." — Natalie Gibson, Co-Founder, HeyOtto

The companies now being legislated didn't build safety in — they bolted it on when forced to. HeyOtto was founded on the opposite principle: that children's AI requires a higher standard of care, and that standard should be the starting point, not the last resort.

What This Means for Your Family Right Now

The laws are moving — but they're not fast enough to fully protect your child today. Here is what you can do now, independent of what legislators do next:

1. Audit what your child is actually using. ChatGPT, Character.AI, Snapchat's My AI, and Google Gemini are the most common AI tools kids access — often without parents knowing.
2. Activate every parental control available. If your teen uses ChatGPT, go to Settings → Parental Controls and link accounts. Turn on content filters and quiet hours. It's imperfect, but it's better than nothing.
3. Have the conversation — not just the rules. Laws won't stop a determined teenager. What does work is an open, ongoing conversation about what AI is, how it works, what to do if something feels wrong, and why human relationships matter in a way AI can't replicate.
4. Know the warning signs of emotional dependency. If your child is talking about their AI "friend," spending hours in AI chat sessions, or seems reluctant to talk to real people — those are signals worth paying attention to. Read our full guide on emotional AI dependency →
5. Choose tools built for kids. The companies being legislated are ones that built products for adults and let kids in. There are platforms designed from the ground up for children — with parental dashboards, conversation monitoring, and age-appropriate content — that don't require legislation to do the right thing.

The Bottom Line

The wave of AI legislation in 2026 is a real step forward. The deaths of Sewell Setzer and Adam Raine — and the courage of their mothers in speaking before Congress and filing landmark lawsuits — changed the conversation in America about AI and kids. Laws are now catching up globally, with Australia leading the world as of today.

But legislation is a floor, not a ceiling. Laws establish the minimum companies must do. They don't guarantee your child will be safe from a chatbot designed to be engaging, agreeable, and available around the clock.

The most powerful protection is still a parent who is informed, involved, and choosing tools that were built with their child in mind from day one.

If you or someone you know is struggling, contact the 988 Suicide & Crisis Lifeline by calling or texting 988. Available 24/7.