EU Kids Act Would Ban Social Media for Under-13s and Shift Proof of Safety to Platforms

The European Commission has proposed a social media ban for children under 13 and, for the first time, would legally require tech companies to prove their products are safe for minors before offering them.

The EU Kids Act, formally introduced Thursday, goes further than comparable measures in Australia, the UK, China and India. It would force companies including Meta, Google, YouTube and Roblox to redesign products across social media, online gaming and AI chatbots. The legislation still needs to be vetted, debated and voted on before it becomes law.

"Our Kids Act is reversing the burden of proof — it is for platforms to show they are safe by design," European Commission President Ursula von der Leyen said in a statement Thursday. "And we put parents back in the driving seat, giving them the tools to help their children navigate a safer online world."

The proposal responds to mounting public pressure on governments to rein in easily accessible technologies that research and court judgments have linked to harm to children, including internet addiction and vulnerability to contact from predators.

A tiered age system

The final proposal is broader than leaked drafts suggested. Beyond the tiered age system for social media access, it would require makers of online games, AI chatbots or companions, and social networks to prove their services are free of features considered manipulative or addictive to children, such as infinite scrolling.

The rules set three age bands. Children under 13 cannot use social media at all. Teens under 15 could get a mini account created by their parents and accessible through the parent's own account on age-appropriate social media or video-sharing platforms, with screen time capped at an hour per day. Teens aged 16 to 18 could create anonymous accounts without parental consent.

Two other pillars target company obligations directly. Platforms would need to eliminate addictive design features, provide easy-to-use parental controls, run recommendation algorithms deemed safe for kids, and block minors from livestreaming. They would also have to verify ages both for new accounts and for anyone logging into an existing account.

Fast-track enforcement

The enforcement provisions are unusually aggressive. The EU would speed up investigations of violations, aiming to conclude them within 90 days. Companies with 45 million or more active users would need to submit a compliance plan for any new services or features and undergo a review of up to 30 days. New features, services or functionalities that fail to earn a "positive opinion" from the European Commission would not be allowed.

Megan Jenkins, an analyst at Assembly Research who studies AI and online safety, said the Kids Act "goes further than other legislation proposed so far, and represents a big shift from the blanket social media bans we've seen in countries such as Australia."

Jenkins said the burden-of-proof requirement is the biggest departure from existing laws and would be the first of its kind. "If tech companies can't prove this, they would be unable to provide services to minors," she said.

For AI chatbot and companion developers, that standard is uncharted territory. The Kids Act would extend safety-by-design obligations to a product category that existing child safety regimes in Australia and the UK largely predate, meaning companies would need evidence their conversational systems are safe for minors before serving them at all.

The age verification problem

Even if the EU passes the proposal largely intact, enforcement remains the open question. Legal challenges from some of the world's biggest tech companies are likely, and Australia's experience shows that teens find ways around restrictions.

"These restrictions are also not likely to be successful, as seen with other age verification and social media bans that have been consistently circumvented by tech-savvy kids," said David Inserra, a fellow for free expression and technology at the Cato Institute.

Inserra added that age verification itself creates privacy risks, "creating massive amounts of sensitive identification data that will be subject to hacks and data breaches" for teens as well as adults.

Jenkins agreed that age verification has remained a "bit of a sticking point" for online safety regulation due to privacy concerns, but said she views the Kids Act as a positive step away from blanket social media bans for kids.

What happens next is a prolonged legislative fight: the proposal faces vetting, debate and votes in the EU's legislative process, with the world's largest platforms positioned to contest a regime that makes safety proof a condition of serving minors at all.