The Pulse
EU Plans Under-15 Limits for AI Chatbots and Social Media
The European Commission plans to propose restrictions on social media, AI chatbots, video platforms and online games for children under 15. The draft EU Kids Act would also require age checks, parental controls and fees from companies to fu
AI.info Team ·
The European Union plans to restrict children under 15 from using social media, video-sharing platforms, AI chatbots and some online games under a proposed EU Kids Act, according to a European Commission document reviewed by Reuters. The plan sets up a clash between Brussels’ child-safety ambitions and the legal and technical work needed to turn a broad age limit into enforceable rules.
The Commission has not yet proposed a final law. Officials are expected to present the initiative on Thursday, September 17, after Commission President Ursula von der Leyen addresses the European Parliament in her State of the Union speech on Wednesday, September 16. The draft still needs to be negotiated with EU member states and the European Parliament before it could take effect.
Reuters’ report, published September 14, 2026, says the Commission declined to comment ahead of the presentation.
Brussels draws a line at 15
The draft would allow children aged 15 and older to create their own accounts. Children between 13 and 14 could receive access to introductory accounts for social media and video-sharing services only with parental approval, limited contacts and strict time limits, according to the document.
Children aged 3 to 12 would be limited to accounts fully controlled by parents and services designed for children under strict safety standards. Children under 3 would have no access to the covered services under the draft. The document describes a graduated system rather than a single rule applied identically to every digital product.
AI chatbots appear alongside social media, video platforms and online games in the proposed under-15 restriction. The document does not provide a separate public explanation of how each age band would apply to chatbots, leaving the treatment of general-purpose assistants and companion-style services to be clarified during the formal proposal and legislative process.
Age checks would become a condition of access
The draft would require social media and video-sharing platforms to verify a child’s age when the user opens a new account. Online gaming platforms would have to perform age verification before users download games. The proposal also calls for companies to provide effective parental-control tools and mechanisms that allow children to report harmful content easily.
Age verification is likely to become one of the most contested parts of the plan. The Commission would need to define how services can check age without creating unnecessary collection of identity information or introducing a system that is easy to evade. The Reuters report does not specify which technical method the Commission would endorse for AI chatbots.
Companies would also face design requirements. The draft calls for measures against addictive designs and harmful feeds, extending the proposal beyond account registration and into how platforms recommend and present content.
Companies would help pay for enforcement
The proposal would require companies covered by the rules to pay a supervisory fee. The money would fund regulators responsible for enforcement and supervision, shifting part of the cost of the new framework onto the services it targets.
That approach would expand the financial and operational obligations already facing major online platforms in the EU. The Digital Services Act requires the largest designated services to assess and reduce risks involving minors, mental well-being and other systemic harms. The European Commission designated ChatGPT as a Very Large Online Search Engine under the DSA on August 31, 2026, giving it four months to comply with additional obligations tied to that designation. The Commission’s designation announcement identifies ChatGPT, Reddit and Roblox as services subject to the bloc’s strictest online-safety tier.
The proposal is not yet a ban
The wording released ahead of the announcement describes a proposed framework, not a rule already binding on platforms or families. Details may change before the Commission presents the initiative, and the final text would have to pass through the EU’s ordinary legislative process.
The distinction matters for AI chatbot providers. Under the current EU AI Act, chatbot operators already face transparency requirements requiring people to be informed when they are interacting with an AI system. The Commission says those obligations became applicable on August 2, 2026, while the new EU Kids Act would address who can access certain services and under what conditions.
The immediate next step is political: von der Leyen and EU tech chief Henna Virkkunen are expected to set out the Commission’s proposal on September 17. After that, the age thresholds, covered services, verification methods and supervisory fees will move from a leaked draft into negotiations among the Commission, national governments and the European Parliament.