EU ChatGPT Search Engine Designation

The European Commission has decided ChatGPT is a search engine. Not a chatbot, not an assistant — a "Very Large Online Search Engine" under the Digital Services Act, the same regulatory bucket as Google and Bing.

The number that got it there: 159.1 million average monthly active users in the EU over the six months ending March 31, 2026. That is more than three times the 45 million threshold that triggers the designation. OpenAI now has four months to assess and reduce the systemic risks tied to how ChatGPT answers questions, the same obligation search engines have carried for years.

Here is the part worth sitting with. The Commission's reasoning is that an AI system which answers a prompt by synthesizing a single confident response carries more accountability risk than a page of ranked links a user has to sort through themselves. Confidence, in other words, is now a regulatory liability.

This will not stay an OpenAI problem. Any assistant that crosses 45 million EU users and answers by searching the web, Gemini, Perplexity, eventually Claude, is reading the same fine print right now. The compliance cost of being useful went up for the whole category.

Do you think regulators are right to treat a confident AI answer as higher risk than a list of search results, or does this formalize a bias against how these tools actually work?