The European Commission formally designated ChatGPT as a Very Large Online Search Engine under the Digital Services Act on Sunday, citing OpenAI's report of 159.1 million average monthly active EU users in the six months to March 2026 — more than triple the 45 million-user threshold that triggers the classification. OpenAI has four months, until the end of 2026, to achieve full DSA compliance.

The immediate regulatory obligations are significant: annual independent audits, mandatory data-sharing with EU authorities and vetted researchers, a public advertising repository, and an obligation to identify and mitigate systemic algorithmic risks. Reddit and Roblox received the same VLOSE designation in the same announcement, though neither raises the same practical questions for the AI visibility field. ChatGPT is the first AI chatbot to be classified by a major regulatory authority as a search engine. That classification, and what follows from it, is the more consequential story.

The DSA's VLOSE framework was designed to make large search platforms legible to the outside world: auditable, researchable, their ranking and recommendation logic at least partially accessible to the regulatory apparatus. That design, applied to ChatGPT, creates a mechanism that practitioners in GEO — Generative Engine Optimisation — have lacked since the discipline emerged. Google Search's ranking logic is partially legible through years of public documentation, algorithm update announcements, Search Console data, and a substantial body of empirical research. ChatGPT's citation logic has, until now, been opaque: practitioners have reverse-engineered it from observed citation patterns, studied it through tools that sample queries, and built probabilistic models based on what appears to work. The DSA obliges OpenAI to make that logic more transparent — through researcher data access, audit findings, and the public advertising repository — by the end of this year.

The GEO implications unfold at two timescales. In the short term — the four-month compliance window — the practical changes are likely modest. OpenAI will implement the minimum infrastructure to satisfy the Commission. But the data-sharing requirement for vetted researchers is genuinely new: for the first time, AI citation research can be conducted with access to ChatGPT's internal data rather than only from the outside. The Citation Lifetime Score methodology that RankCaster AI published last week, the Brainlabs conversion data, the Semrush Reddit citation analysis — all of these were constructed from external sampling. Inside access to ChatGPT's source selection logic, at scale, would transform the empirical basis of the field.

At the longer timescale — one to three years, as DSA enforcement builds precedent — the classification creates a gravitational pull toward transparency that will affect how OpenAI communicates with the practitioner community. Google's detailed documentation of how it evaluates content exists in part because regulatory and legal scrutiny made opacity untenable. The same dynamic, applied to ChatGPT, will eventually produce the equivalent: guidance, documentation, and transparency mechanisms that GEO practitioners can use the way SEOs use Google's published guidance today.

There is a further dimension worth noting. The DSA's definition of a Very Large Online Search Engine turns on user scale and market effect, not on how the platform was designed or what its operators call it. OpenAI has consistently described ChatGPT as an AI assistant, not a search engine. The Commission's classification says that the distinction does not change the function: if 159 million EU users are using it to find information, evaluate sources, and make decisions, it is performing a search function regardless of the interface layer. That framing has implications beyond Europe. It is an argument that any sufficiently large AI chatbot with information retrieval capabilities is, for regulatory purposes, a search engine — and should be treated as one. If the precedent spreads to US regulatory frameworks, the entire posture of the AI lab industry toward transparency and accountability changes.

For CMOs and marketing directors planning AI visibility strategy, the concrete near-term recommendation is watch the researcher data access provisions closely. The first published research using DSA-mandated ChatGPT data will be the most empirically grounded analysis of ChatGPT's citation selection logic the field has seen. That research will be more valuable than any prior reverse-engineering exercise. The window between compliance implementation and the first major research publications — roughly mid-2027 — is the window in which brands that have already built high-CLS content architectures will have the most durable advantage.