The European Union has officially designated OpenAI’s ChatGPT as a “Very Large Online Search Engine,” subjecting the AI chatbot to the same stringent regulatory oversight as Google Search and Microsoft Bing. For European users and global tech companies, this classification under the Digital Services Act (DSA) means OpenAI must now actively mitigate systemic risks, ranging from the spread of illegal content to algorithmic biases, or face severe financial penalties.
Online services trigger these strict legal designations once they surpass 45 million monthly active users within the European Union. While ChatGPT likely crossed this threshold long ago - boasting over 1 billion weekly users globally - the European Commission reportedly struggled with how to categorize the service. Because the DSA was passed in 2022, before generative AI went fully mainstream, the legislation lacked clear parameters for classifying AI-driven conversational agents.
Ultimately, regulators decided to place ChatGPT in the search engine category. Alongside OpenAI's chatbot, the Commission announced that Reddit and Roblox have also crossed the user threshold, earning the designation of “Very Large Online Platforms” (VLOPs). They join a list that already includes major social media networks like Facebook, X, and TikTok, as well as marketplaces like Amazon and Apple’s App Store.
These new designations mean that ChatGPT, Reddit and Roblox will now be held to a higher standard of scrutiny and accountability in the European Union, in line with their large impact on our citizens and society.
- Henna Virkkunen, Executive Vice-President for Tech Sovereignty, Security and Democracy, European Commission
The new designations grant the European Commission expanded authority to oversee and investigate these platforms. ChatGPT, Reddit, and Roblox now have a four-month window to comply with the DSA's enhanced requirements. This includes conducting mandatory assessments of the systemic risks posed by their algorithms, specifically regarding the spread of illegal content, impacts on elections, threats to fundamental rights, and negative effects on the physical and mental well-being of minors.
Failure to comply carries massive financial consequences. Companies that violate the DSA can face fines of up to 6% of their annual worldwide revenue. The Commission has already demonstrated its willingness to enforce these penalties; in December, it fined Elon Musk’s X 120 million euros over deceptive design practices related to its verification badges and a lack of advertising transparency. Regulators have also targeted TikTok, investigating its infinite scroll and highly personalized recommendation systems for potential DSA violations.
A Reddit spokesperson confirmed the company's readiness to adapt, stating that "Reddit is committed to complying with the EU Digital Services Act. We’ve been preparing to meet the VLOP requirements and look forward to coordinating with EU regulators while providing transparency for Reddit users." OpenAI and Roblox have not yet issued public statements regarding their new regulatory status.
The Generative AI Loophole is Closing
The European Commission’s decision to label ChatGPT as a search engine is a pragmatic, albeit slightly awkward, solution to a glaring legislative gap. By forcing a generative AI model into a regulatory box designed for traditional link-retrieval systems like Google, the EU is signaling that it will not wait for new, AI-specific laws to take effect before reining in the technology's massive societal influence.
However, this classification carries profound implications for OpenAI's future liability. Traditional search engines are generally protected as neutral conduits of information, but ChatGPT synthesizes, rewrites, and generates new text based on its training data. If the EU legally treats it as a search engine, regulators may soon demand the same level of transparency regarding source attribution and content demotion that they require from Google.
This sets a dangerous precedent for AI companies operating in Europe. If an AI hallucinates defamatory information or generates illegal content, OpenAI can no longer easily dismiss it as a beta software glitch. Under the DSA, they are now legally obligated to prove they have systemic guardrails in place to prevent it, fundamentally shifting the burden of proof from the user to the algorithm's creator.
