Loading page…
The designation places an AI chatbot in the EU’s top platform-law tier, requiring risk reviews, mitigation and outside scrutiny alongside Reddit and Roblox.
Listen to this story
OpenAI must now treat ChatGPT as an EU-regulated online search engine after the European Commission found it exceeded the 45 million monthly-user threshold. The classification gives OpenAI four months to document systemic-risk assessments and mitigations covering illegal content, minors, well-being, fundamental rights, elections, and public security. It also brings external audits, data-access obligations, and potential Commission investigations or fines.
The designation was announced Aug. 31 and applies to ChatGPT, Reddit, and Roblox under different DSA categories.
ChatGPT is the first AI chatbot placed in the DSA’s heightened regime; Google and Microsoft search services were already covered.
OpenAI’s compliance work must address algorithmic systems and service-wide risks, not only individual chatbot answers.
The European Commission has brought ChatGPT into the EU’s strictest Digital Services Act tier, classifying the chatbot as an online search engine and giving it four months to meet new risk-management duties.
The Aug. 31 designation also covers Reddit and Roblox, but through the online-platform category. All three reported more than 45 million monthly active users in the EU, the threshold for the law’s very-large designation.
The DSA, enacted in 2022, increases obligations as a service’s user base grows. Google and Microsoft search services had already been designated under the framework. The Commission’s decision treats ChatGPT as an information-retrieval service, while Reddit and Roblox are services that publish third-party content. ChatGPT is the first AI chatbot described as subject to these heightened rules.
Within four months, the companies must assess and mitigate systemic risks from their services and algorithmic systems. The required work covers several categories of potential harm, rather than a narrow review of individual posts or answers.
The heightened regime also requires risk-management systems, external audits and data access. The Commission can investigate compliance and impose fines for violations. For OpenAI, the designation therefore creates a defined operational test: show how ChatGPT identifies and reduces the DSA’s specified risks under regulatory scrutiny.
Lena-Maria Boswald of German tech think tank Interface said before the decision that it could set a precedent for regulatory expectations around large generative-AI systems used in the EU. She said OpenAI already faces risk-management obligations under the AI Act, but that the DSA could widen the scope. The immediate question is how the Commission applies platform-law obligations to an AI service after the compliance period ends.
Editorial analysis
Our view: the important choice is the legal route. The Commission did not create a chatbot-specific category; it classified ChatGPT as a search engine and applied an existing regime for large digital services. That gives EU regulators a practical way to examine a generative-AI product’s systemic effects while broader AI rules also apply. The next meaningful evidence will be OpenAI’s compliance approach, especially its risk assessments, mitigation measures, audit arrangements and data access. That scrutiny arrives as ChatGPT advertising has expanded in Europe and account-connected workplace tools have widened the service’s use cases.
Loading discussion...
Be the first to share a perspective or an experience.
Reader comments
Newest comments first. Replies stay oldest first.