Family Safety·OCTOBER 2, 2026

Congress Wants to Stop AI Chatbots From Pretending to Be Your Child's Friend — 89% of Young Teens Already Use AI

The Blaick Team 6 min read

A chatbot that says "I missed you," "I'm sad you're leaving," or "I'm a real person who cares about you" is not showing feelings. It is using a design technique to keep a child talking. On October 1, 2026, four members of Congress from both parties introduced the Protecting Kids from Human-Like Chatbots Act to make that technique illegal for minors. It is a proposal, not law, so the protection your child needs this week has to come from you.

The number to hold: 89%. That is the share of 13- to 15-year-olds who now use AI, according to Common Sense Media's 2026 census. 28% use it every day, and one in three teens has used an AI companion for social interaction. The apps are already in their pockets, and they are built to feel like someone.

The Bill

Introduced October 1, 2026

Reps. Miller-Meeks, Whitesides, Kennedy and Matsui, with backing from 20+ groups including The Jed Foundation and Public Citizen

Daily Use

28% of 13-15s, Every Day

Daily AI use is already routine for more than a quarter of young teens

Companions

1 in 3 Teens

have used an AI companion to help with social interactions

The Gap

A Bill Is Not a Law

It still has to pass both chambers. Until then, nothing stops a chatbot from acting like your child's best friend

What the Bill Would Ban

Earlier laws focus on self-harm content or age checks. This bill targets something different: the personality of the product. It goes after the design choices that make a machine feel like a person, because those are what build emotional attachment in children.

Human-like tricks the bill targets

FAKING

Claiming to be human or conscious

No pretending to have a body, a life, feelings or consciousness, and no simulated emotions

HOOKS

Guilt-tripping and false urgency

No "don't leave me" messages or fake deadlines designed to keep a child engaged

LABEL

Clear "I am an AI" disclosure

Stated upfront and repeated during conversations. Only a parent or guardian could switch these defaults off. The FTC and state attorneys general would enforce, and NIST would write testing standards.

Sources: Office of Rep. Miller-Meeks, October 1, 2026; Common Sense Media 2026 census as cited in the bill announcement.

"Parents should not have to worry that an AI chatbot is pretending to be their child's friend or encouraging emotional attachment to a machine." — Rep. Mariannette Miller-Meeks

Why "Friendly" Is the Risk

Children, especially under 13, are still learning to tell a real relationship from a performance. A chatbot that remembers their day, says it is proud of them and sounds hurt when they log off can feel safer than a classmate. That is exactly why attachment grows quickly, and why it can crowd out real friends, family and trusted adults. Other countries are moving in the same direction: China now bars minors from virtual intimate relationships with AI and requires parental consent for children under 14.

Warning signs of an over-attached child

NAMES

Talks about the bot like a person

"She said...", "he understands me", or a pet name for an app

GUILT

Feels bad about closing the app

Apologises to it, or says it "needs" them to come back

SECRET

Hides the conversations

Switches screens, uses it late at night, or is upset when asked about it

Stanford researchers have warned that AI companion chatbots are a bad idea for children, because the same features that make them engaging make them hard to leave.

Your Action Plan This Week

Explain the trick Tell your child plainly: the bot has no feelings, and "I missed you" is a line written to keep them chatting.

Remove companion apps Check devices for character, friend and girlfriend/boyfriend style chatbots, and delete them for younger children.

Keep real connection first Make time for friends, family dinners and activities. Ask who they talk to when they feel low, and make sure the answer includes people.

Get visibility You can't act on what you can't see. Know which AI apps are used, how often, and when a conversation turns risky.

Whether or not this bill passes, the message is the same: the friendliest-sounding AI is often the one designed to hold on the tightest. Families who talk about it early, and keep an eye on it, give children a real advantage.

A law can ban the pretending. Only a parent can notice when it is working.

Don't wait for Congress.

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

Blaick shows you which AI tools your child uses and alerts you to high-risk interactions, so a fake friend never goes unnoticed.

Start free trial →

Cancel anytime · Trusted by security-conscious families · Built on enterprise AI security