ChatGPT is now a search engine under EU platform law

The most important AI story on Monday, August 31, 2026 is not another model launch.

It is that ChatGPT is no longer just a chatbot in the EU rulebook. The European Commission designated it a Very Large Online Search Engine under the Digital Services Act. That is the same heaviest search-engine tier as Google Search and Bing. For a small team that already ships on ChatGPT or the OpenAI API, model choice and EU go-to-market just became a vendor-risk file.

Yesterday's field note was a music-copyright complaint that turned training-data provenance into a board file. One sentence is enough. Today's file is the EU putting a conversational product in the search-engine bucket.

What the Commission actually designated

I am staying with what the primary sources confirm.

Today the Commission designated ChatGPT as a Very Large Online Search Engine (VLOSE). It designated Reddit and Roblox as Very Large Online Platforms (VLOPs). The services declared that they reach at least 45 million average monthly users in the EU, which is the threshold for this tier. (Commission press note)

After notification, the services have four months to comply with the additional VLOP and VLOSE obligations. The Commission's own wording is "by January 2027." Some secondary coverage says the end of December 2026, and at least one recap says the end of November. I am using the Commission sentence and the four-month clock from notification. (Commission press note)

The extra duties include assessing and mitigating systemic risks from the service and its algorithmic systems. The Commission lists illegal content, negative effects on minors, users' physical and mental well-being, fundamental rights, electoral processes, and public security. The usual VLOSE package also attaches: risk assessment, independent audits, researcher data access, and ranking and advertising transparency. (Commission press note, PPC Land)

The supervision list, updated the same day, names OpenAI Ireland Limited as the provider of ChatGPT and classifies the service as a very large online search engine. It puts ChatGPT at 159.1 million average monthly active users in the EU. Reddit Netherlands B.V. is listed at 57.2 million as a very large online platform. Roblox is listed at 46.6 million as a very large online platform. Those counts are marked as reported by the provider. They are not an independent audit. (Commission supervision list)

ChatGPT now sits in the search-engine row with Google Search, listed at 364 million, and Bing, listed at 119 million. That is the classification that matters. Reddit and Roblox were designated the same morning. They are a different legal object. (Commission supervision list)

Secondary reporting of the Commission's reasoning is consistent enough to use. The Commission describes ChatGPT as a hybrid service, an AI system that engages with and responds to prompts and queries, including by searching the web, which is why it lands as a search engine rather than a classic platform. Executive Vice-President Henna Virkkunen said the designations mean ChatGPT, Reddit, and Roblox "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." I am treating that sentence as reported Commission language, not as a finding about your product. (PPC Land, Euronews)

One piece of soft context belongs in the same folder. OpenAI said on August 18 that ChatGPT Ads would expand the following week to 31 European markets, shown only to Free and Go users. The designation lands while the boundary between an answer and a paid placement is still being drawn. I am not inventing ad-product detail beyond that. (OpenAI)

Why this is a small-team story

You are not going to write OpenAI's systemic-risk assessment. That is not the point.

The usable file is not "ChatGPT is now illegal in Europe." Designation is a supervisory status. It is not a ban, and it is not a fine. OpenAI still has four months on the Commission's clock. Other labs have their own EU files. If you rip ChatGPT out of a product this afternoon because a Monday headline scared you, you have confused a designation with an outage.

What changed is the legal category of a product your customers already treat as search. A conversational assistant that answers by looking things up is now in the same DSA bucket as Google Search. Systemic risk assessment, independent audits, researcher access, and ranking or ad transparency now attach to that product. If you have EU users, or you sell to EU customers who do, "we just wrap ChatGPT" is no longer a complete vendor answer.

The designation names the ChatGPT service under OpenAI Ireland Limited. It does not automatically re-label every OpenAI API SKU in your stack. Do not overclaim that. Do write down which of your workflows actually depend on consumer ChatGPT, which depend on the API, and which already run on another lab. A buyer who asks what DSA exposure sits behind the assistant in your demo will not accept a shrug.

Keep a second provider in the same workflow. A four-month compliance clock, a ranking-transparency change, or a customer who will not accept a newly designated search engine should not strand the product. Yesterday's note was about the rights stack under a model you kept. This one is about the rulebook around the assistant your users already treat as a search box.

My analysis: the category is the load-bearing object

I do not think a small team should pretend it can litigate VLOSE classification. You will not get a better compiler out of that. You can copy the split the Commission just made.

A chatbot is one object.

A search engine is another.

Those can live in the same product. They are not the same legal file. Once the Commission treats prompt-and-answer, including web search, as a search engine, the heavy DSA duties follow the answers, not a page of blue links. Ranking transparency, researcher access, and systemic-risk work now sit on a conversational interface. That is new, even if the underlying model did not change this morning.

If your vendor memo lists price, latency, and a safety page, and has no line for EU designation status, you are not doing procurement. You are doing shopping. Shopping is fine until a customer, an insurer, or a prime asks what you knew about the assistant's legal category and the only answer you have is "it is a chatbot."

I would rather be slightly boring here. Write down which products depend on ChatGPT. Write down whether you have EU users or EU buyers. Write down what the vendor has actually published about the four-month clock, ranking transparency, and where ads sit relative to answers. Update it when the supervision page moves. That is a one-page file. It is also the difference between a headline and a plan.

What I would do if I shipped on ChatGPT this week

I would not yank ChatGPT because a designation landed on a Monday.

I would open the real product list and mark every workflow that cannot survive a sudden ChatGPT change: a ranking or citation change inside answers, a buyer who will not accept a VLOSE in the stack, or an ad-and-answer boundary that shows up in a customer demo. I would turn on a second provider in that same workflow and run last week's tickets through it once, so the fallback is not a weekend fantasy.

I would also ask the vendor question I should have asked when the assistant became load-bearing. What is the written posture on this designation, the four-month clock, ranking transparency, and where sponsored placements sit relative to generated answers? If the answer is only "it is just a chatbot," I would treat that as a gap, not a comfort. I would keep using ChatGPT where it is the right tool. I would stop treating its product category as a substitute for a paper trail.

None of this requires you to pick a side between Brussels and OpenAI. The Commission says the service cleared the 45 million user threshold and now carries the search-engine duties. OpenAI still has to build the file. Both of those sentences can be true enough for a small team to keep shipping while it writes its own page.

If you want help turning that into a real operating setup, start a project conversation.

Sources: European Commission, "Commission designates ChatGPT, Reddit, Roblox under Digital Services Act" (August 31, 2026), European Commission, supervision list of designated VLOPs and VLOSEs (updated August 31, 2026), PPC Land, "ChatGPT faces EU risk rules after declaring 159.1 million users" (August 31, 2026), Euronews, "EU places ChatGPT, Reddit and Roblox under strictest digital safety rules" (August 31, 2026), OpenAI, "ChatGPT Ads expands across Europe" (August 18, 2026)