The European Commission has taken a landmark step in regulating artificial intelligence, officially classifying OpenAI's ChatGPT under the category of platforms with the strictest level of oversight within the Digital Services Act (DSA). This decision automatically imposes a series of stringent obligations on the service regarding content moderation, algorithm transparency, and systemic risk management. Notably, its main competitor—Anthropic's chatbot Claude—did not make it onto this list.
Numbers decide everything: how the entry threshold works
The key criterion for entering the "elite" list is not the quality or safety of the service, but solely the scale of its audience. The threshold is set at 45 million active users per month in European Union countries. Companies self-report their figures, and the regulator makes a decision based on the data provided.
According to OpenAI's latest reporting, ChatGPT's search function is used by approximately 159.1 million people in Europe—more than 3.5 times the established limit. For comparison, Reddit reported 57.2 million users, while Roblox barely crossed the minimum threshold, showing a result just 1.6 million above the required figure.
The mechanism works both ways. In May 2025, the European Commission stripped the adult service Stripchat of its "very large platform" status after its audience fell below the critical threshold.
Why Claude remained in the shadows
Anthropic, like other players, is required to disclose data on its European audience at least twice a year. However, during the reporting period ending October 31, 2025, Claude's figures came in significantly below the established threshold. The company does not disclose the exact number, but it is clear that its European base does not yet reach the status of a systemic player.
That said, this situation could change quickly. Anthropic has not published fresh data for ten months, and during that time the company has announced its plans to go public on the stock exchange. Clearly, new reports could dramatically alter the landscape.
All three platforms now under enhanced oversight are required to bring their systems into compliance with DSA requirements within four months—that is, by January 2027. They will need to conduct assessments of systemic risks related to illegal content, disinformation, users' psychological well-being, and the protection of minors.
The regulator is demonstrating the seriousness of its intentions: in December 2025, the European Commission fined social network X (formerly Twitter) 120 million euros—the first precedent of punishment for violating the DSA. At the same time, the law provides for fines of up to 6% of a company's global turnover.
Claude remains outside the strict oversight zone solely due to its scale, not because of regulator favoritism. Anthropic has already faced a lawsuit over its use of training data, and if the next report shows growth in its European audience, the company will instantly fall under the strictest DSA regime.
My analysis: this situation is a vivid example of how regulatory burden directly depends on business metrics. For the crypto industry and adjacent technology sectors, this is an important signal: the faster your platform grows, the sooner you will face serious compliance requirements. Anthropic, preparing for its IPO, should proactively plan for the scenario of Claude falling under enhanced oversight—otherwise it faces not only administrative costs but also reputational risks.