EU designates ChatGPT under the Digital Services Act

The European Commission has designated OpenAI’s ChatGPT as a Very Large Online Search Engine under the European Union’s Digital Services Act. The decision, reported by The Verge on August 31, 2026, places ChatGPT in the category used for services with at least 45 million average monthly users in the EU. Reddit and Roblox were designated as Very Large Online Platforms in the same announcement.

The designation follows user figures reported by the services themselves. It does not mean that ChatGPT has become a general-purpose search engine in every product context. It means that the Commission is treating the service as a very large online search service for the DSA’s supervisory framework. The distinction matters because the designation activates additional obligations aimed at services with a large potential impact on European users and society.

Under the DSA, ChatGPT must assess and mitigate systemic risks connected to illegal content, the effects of its service and algorithmic systems on minors, users’ physical and mental well-being, fundamental rights, electoral processes and public security. The rules also require greater accountability around how very large services operate. The Commission’s announcement gives the designated services four months after notification to meet the additional obligations, putting the practical compliance deadline at the end of December 2026.

For people using ChatGPT in Europe, the change is mainly about oversight rather than an immediate new feature. Users should expect more attention to safety measures, risk assessments and explanations of how the service affects them. The designation does not itself establish that ChatGPT has violated the DSA, nor does it guarantee that every harmful answer will disappear. It creates a stronger supervisory framework in which the Commission can examine systemic risks and require action.

For OpenAI, the decision adds regulatory work to a service already operating at major scale. Teams will need evidence that safeguards address risks in practice, not only policy documents or isolated examples. For AI makers and companies integrating similar systems, the development is a reminder that reach can bring platform-level responsibilities. The EU is applying a framework designed for large digital services to a conversational AI product, signalling that public impact, user numbers and algorithmic risk increasingly matter alongside model capability.