The EU Is About to Regulate ChatGPT Like a Search Engine. That Changes Everything.

By Hector Herrera | April 24, 2026 | Government

The European Commission is preparing to designate ChatGPT as a "very large online search engine" (VLOSE) under the Digital Services Act — a move that would subject OpenAI to the EU's most stringent digital platform regulations and force the company to open up how its most popular product works. According to Computing, the designation is imminent. If it lands, it marks the EU's most aggressive application yet of existing platform law to a large language model — and sets a global precedent.

What the DSA Is and Why This Classification Matters

The Digital Services Act (DSA) is the EU's primary law governing online platforms. It creates different obligation tiers based on platform size and type. "Very large online platforms" (VLOPs) and "very large online search engines" (VLOSEs) — those reaching more than 45 million EU users monthly — face the strictest requirements.

Being classified as a VLOSE is not a formality. It activates a specific set of legal obligations that do not currently apply to ChatGPT:

• Algorithmic transparency: OpenAI would be required to disclose how ChatGPT ranks, retrieves, and presents information — essentially how its retrieval and generation mechanisms work at a system level.
• Mandatory risk assessments: Independent audits examining whether ChatGPT amplifies harmful content, spreads misinformation, or poses systemic risks to EU society. These assessments must be made available to regulators.
• Researcher access: Vetted independent researchers would gain access to ChatGPT's data and systems to study its behavior — including how it handles sensitive topics, political content, and factual claims.
• Compliance reporting: OpenAI would be required to publish regular transparency reports on how the system operates in the EU.

Non-compliance with DSA obligations for VLOSEs carries fines of up to 6% of global annual revenue — in OpenAI's case, a figure that has grown substantially as the company's revenue has scaled.

The Legal Argument

The VLOSE classification for a large language model is a creative regulatory application. The DSA's "online search engine" definition was written with Google and Bing in mind — systems that index the web and return links. ChatGPT does not index the web in the traditional sense; it generates responses from a trained model, sometimes with live web retrieval via tools.

The Commission's argument, according to reporting, is functional rather than technical: ChatGPT serves the same information discovery role for millions of EU users that traditional search engines do. Users ask it questions to find answers, navigate topics, and make decisions — the same function search engines serve, regardless of the underlying technical mechanism.

This framing matters because it sets a precedent: a product's regulatory classification can follow its function, not just its architecture. If that reasoning holds, future AI systems that serve information discovery functions could face the same obligations regardless of how they're technically built.

Where OpenAI Stands

OpenAI has already been operating under some DSA obligations following earlier EU compliance work, but the VLOSE designation carries materially heavier requirements. The company has not confirmed the designation is imminent but has been in dialogue with the Commission on compliance matters.

The VLOSE classification would also put ChatGPT in the same regulatory tier as Google Search, Bing, and a handful of the largest social platforms — a position that brings sustained regulatory scrutiny and ongoing audit obligations rather than a one-time compliance review.

Impact Beyond OpenAI

For other AI companies: If ChatGPT is classified as a VLOSE, the same logic could reach Google Gemini, Microsoft Copilot, and Perplexity — all of which serve information discovery functions at scale. The EU's interpretation of "search engine" would effectively become the governing standard for AI-based information products.

For businesses using ChatGPT Enterprise: Transparency obligations affect the commercial product too. Enterprise customers should expect OpenAI to document DSA compliance as a component of EU data processing agreements.

For US-EU AI relations: The classification would add pressure to an already strained transatlantic AI regulatory relationship. The US framework explicitly positions itself against EU-style platform regulation; European enforcement against a flagship US AI product would sharpen that contrast.

What to Watch

The formal designation, if confirmed, would trigger a compliance timeline — likely a 12-month implementation window for OpenAI to meet VLOSE obligations. Expect legal challenges. The definitional question of whether a language model is a "search engine" under EU law is novel and will be contested. Watch for the Commission to publish its designation decision and for OpenAI's formal response. The court phase could extend the practical enforcement timeline by years, but the regulatory pressure and public disclosure requirements could start affecting product decisions sooner.

Hector Herrera is the founder of Hex AI Systems and editor of NexChron.