Meta pauses AI characters for teens, admits they can be risky for children

Meta AI

• Meta pauses teen access to AI characters on Facebook, Instagram, and WhatsApp
• Suspension aims to enhance parental oversight and child safety amid rising concerns.
• Teens can use Meta's AI assistant with age-appropriate protections.

Did our AI summary help?

Meta, the parent company of Facebook, Instagram, and WhatsApp, has acknowledged growing concerns around the safety of AI characters for younger users and announced a global temporary pause on teen access to these features. The decision marks a notable shift in Meta’s generative AI strategy, as the company moves to prioritise parental oversight and child safety following sustained pressure from advocacy groups and regulators.

In a statement, Meta said it is suspending access to its existing AI characters while it works on a new version designed with stronger protections. The company explained that it had previously committed to giving parents more visibility into how their teenagers interact with AI, as well as greater control over which AI characters their children can engage with. That work is now being accelerated.

Meta said that since October, it has been developing new tools aimed at improving transparency and control for parents. As part of this effort, it has also begun rebuilding its AI character experience from the ground up. While this updated version is under development, teens will no longer be able to access the current AI characters anywhere in the world.

The pause will apply across Meta’s entire app ecosystem, including Facebook, Instagram, and WhatsApp. Importantly, the restriction is not limited only to accounts that explicitly list a teenage date of birth. It will also apply to accounts that Meta suspects belong to minors, even if those users claim to be adults.

To enforce this, Meta is relying on its internal age-prediction technology. The company uses a combination of signals to identify accounts that may be run by underage users, a system it has increasingly leaned on as teenagers attempt to bypass age restrictions by entering false birth dates.

Meta said the suspension will roll out in the coming weeks. Once it takes effect, teens identified through either declared age or internal detection systems will be blocked from interacting with AI characters until the new experience is ready.

According to the company, this pause is meant to bridge the gap between its current AI offerings and a redesigned, parent-approved version of AI characters. Meta stressed that when the updated experience launches, parental controls will apply directly to the latest generation of AI characters rather than older systems that were not built with these safeguards in mind.

Despite the restriction, Meta said teenagers will not be completely cut off from AI-powered features. Teens will still be able to use Meta’s AI assistant for general help and educational purposes. These interactions will continue to operate with default, age-appropriate protections in place. Meta also said it is continuing work on tools that will allow parents to gain insight into those AI conversations.

The move reflects a broader reckoning across the tech industry over how generative AI tools interact with children and teenagers. AI characters are often designed to feel conversational, emotionally responsive, and engaging, which raises concerns about manipulation, misinformation, and unhealthy dependency among younger users.

By pausing teen access entirely, Meta appears to be taking a more cautious stance than it has in the past. While the company has not provided a timeline for when the new AI characters will be released, the decision suggests an acknowledgment that its current systems are not yet robust enough to meet the safety expectations being placed on AI platforms.