The EU just called ChatGPT a search engine. Your content strategy needs to catch up.

On August 31, 2026, the European Commission did something no regulator had done before: it officially classified a chatbot as a search engine.

ChatGPT is now a Very Large Online Search Engine under the Digital Services Act. Same category as Google and Bing. OpenAI has until January 2027 to comply with the full set of obligations, or face fines up to 6% of global annual turnover.

If you are doing any form of content optimization, GEO, or digital marketing for European audiences, this is not a policy footnote. It is a structural shift in how the EU treats AI-generated search results, and it has direct consequences for how brands should think about visibility in AI answers.

What actually happened

The Commission's reasoning is short: ChatGPT answers user prompts, including by searching the web. That makes it a search engine under the DSA, regardless of whether it looks like one.

OpenAI declared 159.1 million average monthly active recipients for ChatGPT search in the EU over the six months ending March 31, 2026. That is well above the 45 million threshold that triggers the strictest tier of DSA oversight.

The Commission also designated Reddit and Roblox as Very Large Online Platforms the same day. But the ChatGPT classification is the one that matters for content strategy, because it is the first time a conversational AI has been treated as a search engine by a major regulator.

The distinction is not semantic. Under the DSA, a Very Large Online Search Engine has specific obligations that platforms do not: systemic risk assessment, an annual independent audit, data access for vetted researchers, and a public ad repository showing every paid placement with targeting parameters and reach by member state.

Why this matters for content and GEO

Here is what changes and what does not.

What changes

Regulatory legitimacy. ChatGPT is no longer a chatbot that happens to search the web. It is a search engine that also chats. That framing affects how regulators, publishers, and courts think about it. When the EU treats ChatGPT as search, it opens the door to publisher compensation frameworks, transparency requirements, and data access rules that previously only applied to Google and Bing.

Ad transparency. Article 39 of the DSA requires designated search engines to publish a searchable, API-accessible repository of every paid ad. That includes the creative, the advertiser identity, who paid, the main targeting parameters, and reach broken down by member state. If ChatGPT runs ads in your market (and ChatGPT Ads launched across 31 European markets on August 24), those ads are now subject to public disclosure.

Data access for researchers. VLOSEs must provide data access to vetted researchers studying systemic risk. This could, over time, produce the first independent studies on how ChatGPT selects sources and cites content. Right now, citation patterns are a black box. The DSA does not force OpenAI to publish its ranking criteria, but it does create a channel for academic scrutiny.

What does not change

Citation selection stays opaque. The designation does not give publishers a right to be cited, traffic, or payment. It does not require OpenAI to reveal how it picks sources. If you were hoping for a legal lever over ChatGPT's citation behavior, this is not it.

Your GEO fundamentals stay the same. Google's own AI Optimization Guide already said GEO is still SEO. The EU classification does not change the technical optimization playbook: clear structure, direct answers, authoritative sourcing, structured data. What changes is the regulatory environment around the platforms, not the signals that get you cited.

The four-month compliance window is OpenAI's problem, not yours. The systemic risk assessments, audits, and transparency reports land on OpenAI. Nothing in the designation requires changes to your website, content, or markup.

The bigger picture: search engines are no longer defined by their interface

This is the part most coverage misses. The Commission explicitly said the classification is based on what the service does, not what it looks like. ChatGPT does not need to display a page of ranked links to qualify as a search engine.

That is a significant legal precedent. It means any service that retrieves web information, synthesizes it, and presents it to answer a user query can be classified as a search engine. The functional definition is what matters, not the interface.

For content strategists, this means the universe of "search surfaces" just expanded in a legal sense. It is no longer just Google, Bing, and the traditional engines. Any AI system that searches the web and answers questions is, in the EU's eyes, a search engine.

This aligns with what we have been tracking since May. Our data on AI Mode, AI Overviews, and ChatGPT citations showed that these surfaces cite completely different sources. The EU now treats them all as search. Your measurement framework should too.

What to do about it

1. Stop treating ChatGPT as a side channel. If your content strategy still treats AI search as an experimental add-on to SEO, update that. ChatGPT is now legally a search engine in the world's second-largest economy. Its 159 million EU users are not a rounding error.

2. Track ChatGPT referrals separately. We have been saying this since our first data on AI search conversion rates: AI-referred visitors convert at 4x the rate of Google visitors. The EU classification makes this more urgent. If you are not tracking ChatGPT as a distinct traffic source in your analytics, you are blind to a channel that regulators now treat as equivalent to organic search.

3. Watch for the ad repository. When OpenAI publishes its DSA ad repository (required by January 2027), it will be the first public dataset showing exactly what brands are advertising inside AI answers, with targeting and reach data. That is competitive intelligence you will not get anywhere else.

4. Prepare for researcher studies. The data access provisions for vetted researchers will eventually produce independent analysis of ChatGPT's citation patterns. When those studies drop, they will tell you more about how to get cited than any vendor's tool. Keep your content clean and your structured data tight in the meantime.

5. Revisit your three AI search surfaces framework. AI Overviews, AI Mode, and now Information Agents are three distinct surfaces with different citation behaviors. The EU treats all of them as search. Your optimization and measurement should treat them as distinct channels within search, not as one monolithic "AI."

The bottom line

The EU just told the world that a chatbot with web search is a search engine. That is not a theoretical classification. It carries real obligations, real fines, and real consequences for how AI search is regulated going forward.

For content teams, the practical takeaway is simple: optimize for the answer, measure AI referrals as a distinct channel, and stop thinking of ChatGPT as a novelty. It is a search engine now. Act like it.

This analysis is based on the European Commission's official designation (August 31, 2026), the DSA framework, and Mintec's ongoing measurement of AI search citation patterns across Google, ChatGPT, and other AI engines. For our full dataset on how different AI surfaces cite content, see the 40,000-point data breakdown.