The European Commission has moved ChatGPT into the EU’s highest tier of online oversight, classifying the OpenAI service as a Very Large Online Search Engine under the Digital Services Act. The designation places it alongside the biggest digital services operating in the bloc and reflects regulators’ concern that generative AI’s reach now carries platform-sized risks.¹

The decision also covers Reddit and Roblox, which were designated Very Large Online Platforms. The shared threshold is 45 million average monthly EU users — a marker designed to identify services whose decisions, content systems and safety failures can affect large parts of the public.¹

For ChatGPT, the practical consequence is a new compliance clock. OpenAI must address risks tied to minors, users’ mental health and the spread of illegal content, while meeting the wider DSA’s demands for transparency around recommendation systems. The rules also limit targeted advertising to minors and prohibit targeting based on sensitive personal characteristics including religion, ethnicity, sexual orientation and political beliefs.¹

All three newly designated services have until the end of December 2026 to comply. Henna Virkkunen, the Commission’s executive vice-president for tech sovereignty, security and democracy, said the move means ChatGPT, Reddit and Roblox will face “a higher standard of scrutiny and accountability” matching their “large impact on our citizens and society.”¹

The message from Brussels is blunt: ChatGPT may be an AI product, but in Europe it will increasingly be regulated like a major public-facing information service — with the obligations, and potential consequences, that come with that status.