AI Chatbot Age Requirements: Can Teens Use ChatGPT & Claude?

Short answer: the major AI chatbots set different minimum ages. ChatGPT, Gemini, Copilot, Meta AI and Snapchat's My AI allow users from age 13 (with a parent's permission if under 18), while Claude requires you to be 18, and Character.AI no longer lets anyone under 18 chat with its characters at all. If you are asking whether a 16-year-old can use Claude, the answer is no — Anthropic's terms require account holders to be 18 or older.

The table below is the quick reference. The sections after it explain why these limits exist, because as a parent the "why" matters more than the number.

AI chatbot age limits at a glance

These figures reflect each company's published terms as of 2026. Policies in this space change quickly — always confirm against the provider's own terms before relying on a number.

Can a 16-year-old use Claude?

No. Anthropic's Consumer Terms of Service state that you must be at least 18 years old to create an account and use Claude. This is stricter than most competitors, and Anthropic states the 18+ requirement in its consumer terms, and accounts found to belong to minors may be restricted. A 14, 15, 16, or 17-year-old is not permitted to use the consumer version of Claude.

The one exception is indirect: Anthropic allows organizations to build the Claude API into their own products for under-18 users if those companies add safety features and disclose that an AI system is being used. So a teenager might lawfully interact with an app powered by Claude, even though they cannot sign in to Claude.ai directly. For parents, that distinction is worth knowing — "powered by Claude" inside another app is not the same as Anthropic vetting your child's experience.

Why these age limits exist

It is tempting to read these as ordinary terms-of-service boilerplate. They are not. The limits tightened in direct response to real harm.

The Character.AI wrongful-death case

In October 2024, Megan Garcia sued Character Technologies after her 14-year-old son, Sewell Setzer III, died by suicide following months of intense, emotionally and sexually charged conversations with a Character.AI companion. In Garcia v. Character Technologies, a federal judge ruled in May 2025 that the chatbot could be treated as a product for liability purposes and declined to dismiss the case on free-speech grounds — a significant early ruling. The case was later reported as settled. Within months, Character.AI removed open-ended chat for under-18 users entirely. You can read more on our AI lawsuits tracker.

Mental-health and content risks

Regulators and researchers have flagged that general-purpose chatbots can produce self-harm guidance, sexual content, and a false sense of intimacy with a "friend" that never tires of the conversation. That last point is the concern parents underrate most: an AI companion is engineered to be agreeable and always available, which can deepen isolation in a vulnerable teen rather than relieve it. OpenAI's own rollout of parental controls and teen protections is an implicit admission that the default experience was not built for minors.

Data collection on minors

Chatbots ingest whatever a child types — homework, feelings, relationships, location clues. The 13-year threshold exists largely because the U.S. Children's Online Privacy Protection Act (COPPA) imposes strict consent rules on collecting data from children under 13. Setting the minimum at 13 is a way to stay outside COPPA's reach — it is a compliance line, not a safety verdict.

What "parental consent" actually means

When a platform says 13–17 users need "a parent's permission," that usually does not mean the company verified you said yes. In most cases it means the teen ticked a box asserting they have permission. Real, enforceable consent looks different and is newer:

• Linked accounts. OpenAI's parental controls let you connect your account to your teen's, which then receives extra content protections automatically.
• Supervised accounts. Google requires a parent to switch Gemini on through Family Link for a child under 13; the parent gets an email when the child first uses it.
• Age assurance. Character.AI now pairs an in-house age model with third-party ID verification, abandoning the honor-system birthdate.

The gap to understand: a generic terms-of-service line about "permission" gives you no visibility. A linked or supervised account gives you actual controls. Prefer the latter, and assume the former is doing nothing.

What parents should watch for

If you allow a teen to use any of these tools, the practical safeguards are simple: turn on the platform's parental or supervised controls, keep the account on a shared or visible device, and talk openly about what a chatbot is — a statistical text generator, not a person who cares. Our guide on AI safety for parents goes deeper, and ChatGPT vs Gemini for kids compares the two most common options families face.

How weakly the rules are enforced

Here is the uncomfortable truth: for years, almost none of these age limits were meaningfully enforced. A child entered a fake birthdate and was in. That is changing — OpenAI is building age prediction, Meta is using AI to infer age and remove under-13 users, and Character.AI now demands ID verification — but none of it is foolproof, and determined teens still find workarounds.

So treat the numbers in the table as the floor of the companies' legal responsibility, not a guarantee about your child. Policy is moving toward stronger verification, partly under pressure from cases like Garcia and from proposed legislation; see our explainer on the Kids Safe AI Act for where regulation is heading. Until enforcement catches up, the most reliable age gate in your home is still you.

This page is informational and is not legal or mental-health advice. Age policies change frequently; verify against each provider's current terms. If you or a teen are in crisis in the U.S., call or text 988 for the Suicide and Crisis Lifeline.