EU designates ChatGPT a very large online search engine under DSA rules

The European Commission has designated ChatGPT a Very Large Online Search Engine under the Digital Services Act, according to The Verge. The classification puts OpenAI under some of the EU's heaviest obligations for online services, making the company accountable for mitigating risks tied to ChatGPT's impact on minors, on user mental health, and on the spread of illegal content inside the bloc.

How the designation works

The Digital Services Act is the EU's rulebook for large online platforms and search engines, and its strictest tier applies once a service averages at least 45 million monthly users in the union. According to The Verge, ChatGPT has now crossed that threshold, and the Commission paired the announcement with two further designations: Reddit and Roblox are classified as Very Large Online Platforms and fall under the same regime.

Henna Virkkunen, the EU's executive vice-president for tech sovereignty, security and democracy, said in the announcement that the three services "will now be held to a higher standard of scrutiny and accountability in the European Union, in line with their large impact on our citizens and society."

The choice of category is worth noting. Rather than treating the chatbot as a social platform, Brussels placed ChatGPT in the DSA's search engine track, a framing that suggests regulators see an AI assistant answering people's questions as functionally closer to search than to a feed-based network.

The obligations that follow

According to The Verge, the DSA restricts the designated services from targeting advertising at minors and from building ad targeting around sensitive traits such as a person's sexual orientation, religion, ethnicity or political beliefs. The law also requires greater transparency about how recommendation algorithms operate, pushing the companies to explain more of the machinery that shapes what users see.

For OpenAI, the designation turns recurring criticism about chatbot safety into defined legal duties: mitigating potential harms to children, limiting negative effects on users' mental wellbeing, and curbing the distribution of illegal content through its service.

A deadline at the end of 2026

The Verge reports that ChatGPT, Reddit and Roblox have until the end of December 2026 to comply with the rules, leaving only a few months between the announcement and the compliance date.

Why it matters

The decision extends the DSA's toughest enforcement tier to a generative AI assistant, treating a chatbot used by tens of millions of Europeans as infrastructure on par with the largest platforms and search engines.

That matters for three reasons. First, OpenAI's responsibilities around minors, mental health and illegal content stop being voluntary commitments and become enforceable obligations in a major market. Second, the search engine classification offers an early signal of how Brussels intends to categorise conversational AI services as their reach grows, a question other AI providers will be watching closely. Third, with Reddit and Roblox designated on the same day, the Commission is showing it will apply platform-style duties to any service with sufficient scale, whatever its format — social discussion, gaming or AI chat.