Summary: On August 31, 2026, the European Commission designated ChatGPT as a Very Large Online Search Engine under the DSA, the first AI chatbot placed in that tier. The tier brings systemic-risk assessments, independent audits, and transparency reporting. Reddit and Roblox were designated Very Large Online Platforms in the same wave.
On August 31, 2026, the European Commission handed down a designation: under the Digital Services Act, ChatGPT now sits in the Very Large Online Search Engine tier — the strictest level of regulation the DSA has. No AI chatbot had been placed there before. Reddit and Roblox, named in the same wave, were designated Very Large Online Platforms.
What the VLOSE Tier Carries
The DSA scales its rules to size and risk. Once a service crosses the threshold into the “very large” tier, the obligations step up sharply — systemic risk assessments, independent audits, and transparency reporting are what that level is known for. Put plainly, ChatGPT’s EU operations will from now on run under the heaviest rulebook in the platform economy.
For OpenAI, compliance shifts position: once a concern at the edges of the product, it now sits at the center. Data governance, risk assessment processes, and disclosure documents stop being a one-time public-relations exercise and become a lasting operational burden.
The Chatbot Sorted Into the Search Engine Class
The category is the part of this designation worth turning over. The DSA’s framework was assembled around service classes that already existed — search engines and online platforms — whereas ChatGPT takes the form of conversation. By designating it a VLOSE, the Commission has in effect ruled that a service through which people obtain information by dialogue is, functionally, a search gateway; how the interface looks does not settle the classification.
That ruling anchors the question. Any AI service with a conversational interface that in substance acts as an information entry point will struggle to argue it is “just a chat tool.” The measuring stick has been set down: function, not packaging.
It also settles something the industry had left open ever since chatbots began absorbing queries that used to go to search boxes: which rulebook governs a service shaped like a category regulators never anticipated. The Commission’s answer is pragmatic — fit the new service onto the function the older law already names. Expect future design decisions to be read through the same lens; a conversational wrapper will not move a service into a lighter regulatory tier.
Reddit and Roblox Designated in the Same Wave
The same announcement designated Reddit and Roblox as VLOPs. A community forum and a game platform could hardly be less alike, yet both were recognized as online services with systemic influence. Bundling all three cases into one batch signals that the Commission is systematically extending the DSA’s reach to large online services sitting outside the traditional definitions.
Reading the three names side by side sharpens the point: an AI chatbot, a community forum, and a game platform now receive the same regulatory treatment. The Commission clearly will not let service form drive classification — it is watching where the influence lands.
What This Means for OpenAI and the Industry
- Compliance costs become concrete: the strictest tier’s obligations are a long-term program reaching into product, data governance, and legal work — not a one-off remediation
- Classification risk spreads: the same stick will be held up to other conversational AI services, narrowing the room for regulatory arbitrage
- The user-facing effect is indirect but real: more risk assessment and transparency disclosure puts the system’s operating logic out for scrutiny
The timing deserves attention next to the EU’s AI-specific rulebook. The AI Act keeps its own milestones, but the DSA designation shows the EU is not waiting for AI-dedicated law to reach every question — existing platform regulation is being applied to AI services as they stand today. Companies that wrote the DSA off as legacy platform law with nothing to say about a chatbot now have a dated counterexample.
Seen over a longer arc, this is a signal that settled platform law is reaching back to take charge of AI services, running alongside the AI-specific rulebooks that came first. For AI product teams operating in the EU market, the DSA has stopped being a “study later” item.