ChatGPT Named EU's First Very Large AI Search Engine

The European Commission notified three companies on August 31, 2026, that they had received new designations under the Digital Services Act (DSA): ChatGPT, made by OpenAI, became a very large online search engine, while Reddit and Roblox became very large online platforms. ChatGPT is now the first generative chatbot in the world to fall into that category, the strictest tier of oversight that European law applies to internet services.

The threshold for the designation is simple: any service that averages at least 45 million monthly users in the European Union must report that figure to the Commission every six months, and the Commission then decides whether to apply the label. For search within ChatGPT, OpenAI reported roughly 159 million average monthly users in the Union, measured over the six months ending in March 2026. Reddit reported 57.2 million users, and Roblox reported about 46.6 million (one source cites a rougher estimate of around 48 million). With these three additions, the number of companies designated as very large platforms or search engines has climbed to 28, a list that already included Google Search and Microsoft’s Bing.

Why ChatGPT was harder to classify than Reddit or Roblox

The Digital Services Act was written before ChatGPT existed, so the law recognizes platforms that host other people’s content and search engines that answer queries, but not a chatbot that does both. The Commission ultimately placed ChatGPT in the search engine category because it answers user queries, including by searching the internet. Reddit and Roblox were more straightforward cases: both services let users share other people’s content publicly, which is the legal definition of a platform.

Henna Virkkunen, the Commission’s executive vice president for tech sovereignty, security and democracy, explained the decision in a statement:

“These new designations mean that ChatGPT, Reddit and Roblox will now be subject to a higher standard of oversight and accountability in the European Union, in proportion to their major impact on our citizens and on society.”

The three services now have four months to meet a set of additional obligations, putting the deadline at the end of December 2026 (one American source puts it at the end of January 2027, while European sources and the Commission itself describe it as four months from notification). The central obligation is a written assessment of the service’s own systemic risks: how much it could contribute to the spread of illegal content, its effect on minors, on users’ physical and mental health, on fundamental rights, on electoral processes and on public safety. Alongside the assessment comes an obligation to mitigate those risks, including by changing product design or the recommendation algorithm. The rules also add an annual independent audit, a publicly searchable archive of every ad along with data on the buyer and the targeting used, data access for vetted academic researchers, and at least one algorithmic feed that does not sort content by user profile.

The Commission shares oversight of ChatGPT and Reddit with the Irish regulator Coimisiún na Meán, and oversight of Roblox with the Dutch Authority for Consumers and Markets. Penalties for non-compliance can reach 6 percent of a company’s worldwide annual turnover, and the Commission has already shown it will use that ceiling in practice: on December 5, 2025, it fined the network X 120 million euros for concealing code that affected content visibility.

All three companies responded with statements. An OpenAI spokesperson said that “search in ChatGPT functions as a search service under the Digital Services Act” and that the company “is preparing to meet the additional compliance obligations that come with the very large online search engine designation.” Roblox’s statement was longer:

“We’re proud to be the first gaming platform to reach this milestone in the European Union, and that growth was driven by our community of creators and our commitment to a safe, civil and age-appropriate experience online.”

A Reddit spokesperson added that the company “has been preparing to meet the obligations of a very large online platform” and “looks forward to working with EU regulators with transparency toward Reddit’s users.”

What the designation does not mean is that the European Union now oversees what ChatGPT is or isn’t allowed to tell its users. The obligation OpenAI has taken on is to write, publish and carry out a risk assessment of its product, not to alter the chatbot’s answers on a regulator’s instruction. The distinction matters: oversight of search engines has traditionally looked at how other people’s links get ranked, while a ChatGPT user receives a single finished answer with no view of the alternatives. That means the risk assessment now has to cover exactly that step, the one where judgment has shifted from the reader to the model.