ChatGPT Is Now Officially Designated as a Search Engine

On 31 August 2026 the European Commission designated ChatGPT as a Very Large Online Search Engine under the Digital Services Act. This makes ChatGPT the first AI chatbot to be classified this way. It was named alongside Reddit and Roblox, which are both designated as Very Large Online Platforms. The classification brings OpenAI’s conversational interface under the same transparency and accountability framework that already governs Google and Bing. It changes how organisations need to think about SEO strategy now that a synthesised answer sits under the same duties as a ranked result.

ChatGPT, Reddit and Roblox all cleared the Act’s threshold of 45 million average monthly users in the EU. According to the European Commission, ChatGPT qualified as a search engine. The justification behind this is that it answers user prompts and performs live web searches. OpenAI reported around 159 million average monthly EU users for the six months to March 2026. It has until the end of 2026 to meet the new obligations. There’s a compelling argument for them to comply, as penalties for the most serious breaches reach up to six percent of global turnover. For organisations working across European markets or serving international audiences, the ruling adds a compliance layer that sits alongside existing accessibility and data protection duties.

What the DSA Classification Means for AI Search

The Digital Services Act asks its largest search engines to be transparent about how content is ranked, recommended or shown to users. For traditional search that means disclosing ranking signals and how they are weighted. For ChatGPT it means openness about how training data, source selection and answer synthesis decide what a user sees. The same threshold applies to both, the 45 million monthly EU users that marks out the strictest tier of the Act.

From here OpenAI carries the obligations the Act places on the biggest platforms. It has to give users clear information about the sources behind a generated answer. It should also let them contest a response that’s inaccurate or harmful. Systems should be put in place to stop illegal content and disinformation being amplified. Beyond that, the Act requires annual independent audits of these systems and regular assessment of the risks they pose to society. This is the same standard already met by Google, Meta and the other major platforms. For any organisation whose content feeds those answers or appears among the cited sources, the stakes around accuracy and attribution rise accordingly.

Implications for UK Organisations and Public Sector Bodies

Public sector bodies working under the Public Sector Bodies Accessibility Regulations 2018 feel this the most. Their content already meets accessibility standards and would now face the transparency expectations the DSA sets for services reaching the EU. NHS Digital guidance already asks for content that’s accurate, clearly sourced and reviewed regularly for clinical and operational correctness. The designation extends that thinking to answers assembled by AI. Several sources can be blended together with little attribution – unless the platform cites them properly.

Organisations in regulated sectors such as healthcare, financial services and professional services now work on two fronts at once. They need content that’s structured to rank in traditional results through sound technical SEO, as well as content that’s formatted so an AI system can summarise and attribute it correctly. The second of those calls for citations and structured data. Those implementations let a language model pull out facts without misattributing them or inventing detail.

How the UK Regulatory Environment May Respond

Westminster has not said whether it will bring in equivalent law. The signals point one way though. The ICO’s guidance on AI and data protection already expects any organisation running AI systems to be transparent about its data sources and its decision logic. Ofcom’s oversight of online platforms keeps widening. Applying those principles to search tools that generate answers – not only rank pages – would put the UK on the same footing as the EU. This would give businesses trading across both a single standard to hold to.

For businesses based in the UK, there’s an immediate question. Will content already tuned for Google’s AI Overviews and traditional search clear a higher transparency bar if the UK adopts rules like the DSA? Thought leadership, case studies, technical documentation and service pages that surface in AI answers may need firmer sourcing behind their claims. They’ll likely also need clearer factual statements the systems can rely on. This is a question of content strategy, not only search tactics.

What This Means for Search Visibility Strategy

The designation settles something that was already becoming clear. In regulatory terms, there’s no longer a meaningful line between a ranked result and a synthesised answer. For anyone running an AI SEO programme, that puts the emphasis on content a model can verify, attribute to a named source and read without tripping over ambiguity.

Organisations now have to prepare for a setting where answers generated by AI are held to the same transparency and accuracy standards as traditional rankings. This asks for tighter governance over content quality and sourcing.

The parallel with accessibility is close. The W3C’s WCAG 2.1 guidelines ask for content that’s perceivable, operable, understandable and robust. Much of that carries straight over to AI search. Content has to be structured for machines to parse, marked up with schema and free of the ambiguity that leads to synthesis errors. Organisations already bound by accessibility rules start from a stronger position than those that treated AI search as an afterthought.

Preparing for Regulatory Alignment

The sensible move is to read the EU designation as a sign of where UK regulation is heading. The ICO’s AI work, Ofcom’s growing remit and the government’s own AI white paper all point towards closer oversight of how algorithmic systems surface and present information. Organisations that build transparency, sourcing and factual accuracy into how they govern content now will have far less to do later if the UK follows.

For marketing and digital teams, the practical work is familiar. It means structured data, clear attribution and content written in plain English. The move from SEO built around rankings to AI search built around synthesis is no longer optional for anyone serving European markets or expecting the UK to align. Visibility and compliance now rest on the same thing. Content a machine can parse accurately and a reader can check independently.