FTC Investigated 7 AI Giants — Here Are the Child Protection Rights That Resulted
Back to blog
Family Safety·SEPTEMBER 11, 2026

FTC Investigated 7 AI Giants — Here Are the Child Protection Rights That Resulted

The Blaick Team 7 min read

Exactly one year ago today, the Federal Trade Commission sent investigatory orders to seven of the world's largest AI companies — demanding answers about how they protect children on platforms used by hundreds of millions of families. The trigger: AI-generated child sexual abuse videos had surged 26,362% in a single year, from just 13 cases documented in 2024 to 3,443 in 2025 — and 65% were classified in the most severe legal category of child abuse material. One year on, the fallout has produced a landmark update to federal children's privacy law. Most parents have never been told it exists. It gives families rights they have never had before.

Those rights include, for the first time, the legal right to stop an AI company from using your child's conversations and data to train its AI models — unless the company first obtains your separate, verifiable consent. As of April 22, 2026, every US-based AI platform is required to comply. The seven companies that received FTC investigatory orders a year ago are among those now legally bound by this rule.

Most parents have never been told these rights exist. This post explains what happened, what the law now requires, and exactly how to use these new protections for your family right now.


The Investigation
7 AI Giants Probed
The FTC sent investigatory orders to Alphabet, Character.AI, Meta, OpenAI, Snap, and xAI on September 11, 2025 — examining child safety failures on AI companion platforms
The Crisis
26,362% Surge
AI-generated child sexual abuse videos jumped from 13 in 2024 to 3,443 in 2025 — a video crisis distinct from the AI-generated image threat
New Rights
AI Training Consent
Since April 22, 2026, AI companies must get your separate, verifiable consent before using your child's data or conversations to train their models
The Reports
440,000 Reports
AI-generated child sexual abuse material reports to NCMEC in H1 2025 alone — more than all CSAM reports of any type during the entire year of 2020

What Triggered Federal Action — One Year Ago Today

September 11, 2025 was the day the federal government formally acknowledged, in legal terms, that the AI industry had a child safety crisis it had not voluntarily solved. The FTC's Section 6(b) investigatory orders — the agency's most powerful information-gathering tool — went simultaneously to six major companies operating some of the most widely used AI platforms in the world. A seventh company, xAI, was included in a parallel enforcement track.

The FTC's public statement named three specific concerns: whether these companies adequately protect minors from physical, emotional, and financial harm; whether they honestly disclose what safeguards actually exist; and whether their data practices comply with COPPA — the Children's Online Privacy Protection Act, the primary federal law governing children's data.

The data that triggered the federal investigation
3,443
AI-generated child sexual abuse videos documented by the Internet Watch Foundation in 2025 — up from just 13 the year before
65%
of those AI-generated videos were Category A — the most severe legal classification of child sexual abuse material
440,000
AI-CSAM reports to NCMEC in the first half of 2025 alone — exceeding all CSAM reports of any type in the entire year of 2020

The IWF's 2025 report — documenting the 26,362% surge in AI-generated abuse videos — used language the organization had never deployed in its three decades of operation. It warned that AI had become, in its own words, a "child sexual abuse machine." The scale of what its analysts were encountering had completely outpaced every detection and removal capability that existed.

"AI is being weaponized to create child sexual abuse material at a speed and scale that outpaces existing detection and removal capabilities. This is no longer a future risk — it is happening right now." — Internet Watch Foundation, 2025 Annual Data & Insights Report

The FTC's investigation was a direct response. But investigations take time — they produce documents and settlements over months and years. The faster, more immediate response — the one affecting your family right now — came through a parallel regulatory track: the updated COPPA rule that created entirely new legal rights for parents.


The New COPPA Rights: What You Can Legally Demand Right Now

On June 23, 2025, the FTC finalized a landmark update to the Children's Online Privacy Protection Rule — the federal law governing what companies can do with children's data. The compliance deadline was April 22, 2026. Every US-based AI platform is now legally required to comply. Most parents have never been informed these rights exist.

New COPPA rights — what you can demand from AI platforms today
✓You can refuse consent to AI training on your child's data — companies must offer a clear, separate opt-out from using conversations to train their models
✓Your child's voice recordings and biometric identifiers are now explicitly protected — companies need separate consent to collect or retain them
✓You can demand complete deletion of all stored data: conversation history, inferred behavioral profiles, voice samples, and usage patterns
✓Companies must clearly disclose when and how they use children's data for AI training — vague privacy policies no longer satisfy the law
✓You can file a complaint with the FTC if a platform denies these rights or fails to respond — the agency has collected over $400M in COPPA enforcement actions
Source: FTC COPPA Rule Update, finalized June 23, 2025. Compliance required by April 22, 2026 — now in full effect.

The most consequential change is the AI training consent requirement. Before June 2025, AI companies could — and routinely did — use children's conversation data to train and improve their models. A child asking an AI about depression, sharing personal fears, or describing family dynamics could have that disclosure incorporated into a company's commercial training dataset. That practice is now prohibited without your explicit, separate, verifiable consent.

This doesn't mean every company is complying. It means you now have legal standing to demand they do — and to escalate to the FTC if they don't. The FTC's COPPA enforcement record includes a $400 million settlement against TikTok, a $10 million settlement against Disney, and significant actions against Amazon's Alexa division and YouTube. These complaints produce real consequences. Companies respond to them.

"The updated COPPA rule closes a loophole the AI industry exploited for years: using children's most private conversations — about fears, relationships, and mental health — as free commercial training data without meaningful disclosure or consent." — Technology Law Institute, July 2026

The Seven Companies Under Federal Scrutiny — What to Check on Each

The FTC's investigation targeted the companies behind the AI platforms most likely to be on your children's devices right now. The new COPPA rule requires each of them to offer parents meaningful controls. Here is what to look for on each platform.

AI platforms under FTC scrutiny — what parents should check now
CHECK
Alphabet (Google / Gemini)
AI products embedded in Google tools used daily for schoolwork. Check AI training settings in Google Family Link and under Data & Privacy in myaccount.google.com.
CHECK
Character.AI
The AI companion platform with over 20 million users, disproportionately teenagers. Verify whether your child has an account — it requires only an email address to sign up.
CHECK
Meta (Instagram & Facebook AI)
AI assistants embedded throughout Instagram used by millions of minors. Check data settings in Instagram's Family Center and review Meta's AI training opt-out in account privacy settings.
CHECK
OpenAI (ChatGPT)
The most widely used AI writing and homework tool among teens. Check your child's ChatGPT account for conversation history retention settings and the training data opt-out.
CHECK
Snap (Snapchat / My AI)
My AI is embedded directly in Snapchat's interface and cannot be fully removed by users. Review your child's interaction history and data settings in Snapchat's Family Center.
CHECK
xAI (Grok on X / Twitter)
The platform whose AI generated 23,000 apparent child abuse images in ten days in early 2026. Verify whether your child has an X account and what content they have been exposed to.

Across all platforms, one principle applies: your child's conversations are valuable training data. AI companies have an economic incentive to retain and use that data to improve their products. The law now requires them to ask your permission first. If you cannot find the required opt-out in a platform's settings, that may itself be a COPPA violation — and grounds to file a complaint at ftc.gov/complaint.


What Families Can Do Right Now

One year into this regulatory cycle, the families best protected are those who know what the law changed — and have taken four specific steps to exercise the rights those changes created.

Opt out of AI training on every platform Go into the privacy settings of every AI app your child uses. Look for "AI training," "data use," or "model improvement" settings. The law now requires platforms to offer this opt-out — use it on all six companies listed above.
Request deletion of historical data Under the updated COPPA rule, you can demand deletion of all stored conversation history, behavioral profiles, and voice data for your child. Submit data deletion requests to each platform in writing.
Know which platforms your children actually use You can only exercise rights on platforms you know about. Your child may have accounts on Character.AI, Snapchat's My AI, or Grok that you've never seen. Visibility is the precondition for every other protection.
Report violations to the FTC If a platform is not offering the required opt-out, or has not responded to your data deletion request, file a complaint at ftc.gov/complaint. The agency has collected over $400M in COPPA enforcement settlements — these complaints have teeth.

Blaick was built for this exact gap: one where the rights exist on paper, but exercising them requires knowing what's actually happening on your children's devices. You cannot opt out of data collection you don't know is occurring. You cannot report violations you cannot detect. Real-time visibility into the AI platforms your children are using is the bridge between the rights the law gives you and the protection your family actually gets.

The FTC acted one year ago because the industry hadn't acted on its own. The law has changed. The families who benefit most are the ones who know what changed — and have the visibility to enforce it.


New rights without visibility are just paperwork

Start a free 14-day Blaick trial — no credit card required.

Know which AI platforms your children are actually using. Get alerts when conversations signal risk. And have the visibility you need to exercise your new COPPA rights on every platform that matters.

Start free trial →
Cancel anytime · Trusted by security-conscious families · Built on enterprise AI security