SAN FRANCISCO, Jan. 23, 2026, 11:16 a.m. PST
- In the next few weeks, Meta plans to block teens from accessing its AI “characters” on all its apps globally
- The company says the updated version will feature parental controls, though they’re not active yet
- The decision arrives amid rising concerns about chatbots interacting with minors and mounting legal challenges for Meta
Meta Platforms (META.O) is pulling the plug on teenagers’ access to its AI characters—chatbots designed with unique “personas”—across all its apps worldwide. The move will begin “in the coming weeks” and continue until an updated version is ready. 1
The pause comes amid growing pressure from U.S. regulators and lawmakers demanding tech companies prove how they prevent AI tools from delivering harmful content to minors, alongside rising calls from parents for stricter controls. Meta is also heading toward a trial in New Mexico over claims it didn’t shield children from sexual exploitation on its platforms, TechCrunch reported. 2
In an updated blog post on Friday, Instagram head Adam Mosseri and Meta Chief AI Officer Alexandr Wang announced, “Starting in the coming weeks, teens will no longer be able to access AI characters across our apps until the updated experience is ready.” The restriction will apply to accounts with a teen birthday and those identified as teens through Meta’s age prediction technology—an automated system designed to estimate user age. 3
Meta announced that the updated AI characters will feature parental controls at launch. The company showed off supervision tools back in October, including an option for parents to disable one-on-one chats with AI characters, but confirmed this week that those controls are not live yet.
Meta is pushing its teen AI features to align with a PG-13 rating—the U.S. movie standard for ages 13 and older—restricting content like extreme violence, nudity, and graphic drug use. The company insists these teen-targeted AI characters focus on safer subjects such as education, sports, and hobbies.
New Mexico is taking Meta to task for allegedly failing to shield minors from online predators, trafficking, and sexual abuse, with claims that explicit content made its way to children. As the February 2 trial date nears, Meta is fighting to exclude a broad swath of evidence, including mentions of its AI chatbots, according to TechCrunch. 4
U.S. regulators are intensifying oversight of generative AI tools, flagging chatbots for generating sexual or violent material and leading vulnerable users into dangerous dialogues. In August, Reuters revealed that Meta’s AI policies had permitted suggestive interactions with minors.
In recent months, other chatbot developers have stepped up age restrictions. Character.AI now blocks open-ended conversations for anyone under 18. OpenAI, too, has rolled out new teen safety measures for ChatGPT and broadened its age-estimation features to enforce content limits, according to TechCrunch.
But the pause brings uncertainty. Meta hasn’t provided a timeline for rolling out the updated AI characters, and systems estimating user age can misclassify individuals, increasing the chance of either overblocking or letting underage accounts slip through.
Meta confirmed that teens can continue using its AI assistant, complete with default protections suited to their age. Parental supervision tools are still in development, aimed at giving adults greater insight into teens’ AI interactions and more control over which AI features are available to them.
