EU readies age checks and agent rules

Brussels said on April 15 that its European Union age-verification app is technically ready and will be available for citizens to use soon. (politico.eu) European Commission President Ursula von der Leyen announced the rollout at a press conference on Wednesday, calling for a Europe-wide technical solution to protect children online. The Commission had already published a prototype and child-protection guidelines on July 14, 2025 under the Digital Services Act. (ec.europa.eu) (digital-strategy.ec.europa.eu) The app is designed to let a user prove they are over 18 without revealing their exact age or identity. The Commission says the proof is anonymous, the link to the proof provider is cut after issuance, and online services can verify only whether the age claim is valid. (digital-strategy.ec.europa.eu) That system has been in pilot testing since July 2025, with Denmark, France, Greece, Italy and Spain customizing national versions. The Commission says the software is open source and built to work with the European Digital Identity Wallet due by the end of 2026. (digital-strategy.ec.europa.eu 1) (digital-strategy.ec.europa.eu 2) The age-check push sits inside the European Union’s Digital Services Act, which gives Brussels leverage over large platforms’ handling of risks to minors. Last July, the Commission also said member states could still impose their own national social-media bans for minors under the same framework. (digital-strategy.ec.europa.eu) (politico.eu) A separate Commission clarification is moving in parallel on artificial intelligence agents, the software tools that can take steps on a user’s behalf instead of only answering prompts. The European Union’s new AI Act regulates artificial intelligence systems and models through one framework for providers and deployers across the bloc. (ai-act-service-desk.ec.europa.eu) (digital-strategy.ec.europa.eu) The Commission’s AI Act Service Desk now lists a specific question asking how artificial intelligence agents are addressed within the law, signaling that agents are being treated as part of the Act’s scope rather than as a separate legal category. The Commission’s public guidance also says the law is meant to be “future-proof” and applies to artificial intelligence systems and models, not just today’s named products. (ai-act-service-desk.ec.europa.eu) (digital-strategy.ec.europa.eu) That matters for companies building agent-style products on top of general-purpose models such as large language models. The Commission says downstream system providers have their own obligations under the AI Act, while the AI Office directly supervises providers of general-purpose models and, in some cases, systems built on those models by the same provider. (digital-strategy.ec.europa.eu) (ai-act-service-desk.ec.europa.eu) Critics of online age checks have warned that broad verification mandates can create privacy and security risks if badly designed. Computer scientists told Politico last month that some age-check systems are “dangerous and socially unacceptable,” while the Commission is arguing its own design minimizes data collection and tracking. (politico.eu) (digital-strategy.ec.europa.eu) The result is a more concrete European Union rulebook on two fronts at once: checking whether a user is old enough, and deciding when an autonomous software tool counts as regulated artificial intelligence. Brussels now has a prototype app nearing rollout and a public compliance channel that is starting to answer the agent question in plain terms. (politico.eu) (ai-act-service-desk.ec.europa.eu)