ChatGPT is about to become the first AI chatbot to face Europe’s toughest platform rules. The European Commission plans to place OpenAI’s chatbot and the gaming platform Roblox under the strictest tier of the Digital Services Act. The designation could come as soon as August.
The move was first reported by Bloomberg, citing a person familiar with the plan. ChatGPT’s search function would be designated a “very large online search engine.” Roblox would become a “very large online platform.” Both crossed the DSA’s trigger of 45 million monthly users in the bloc.
The Commission was more guarded in public. Its spokesperson, Thomas Regnier, told Reuters a designation was “definitely possible” and could “come sooner or later.” OpenAI said it continues to engage positively with Brussels. Roblox did not comment.
What the label actually demands #
The tag is not a trophy. It adds ChatGPT and Roblox to a list of around two dozen platforms carrying the DSA’s heaviest obligations. That group already runs from Meta and X to Amazon and TikTok. They get four months from designation to comply.
That means annual risk assessments and independent audits, transparency reports, ad-transparency rules, and a recommendation feed not based on profiling. They must share data with regulators and vetted researchers, and report criminal offences. There is also a supervisory fee. Breaching the rules risks fines of up to 6 per cent of global annual turnover.
The point is systemic risk. Designated services must identify and mitigate harms tied to illegal content, elections, discrimination, gender-based violence, and users’ mental and physical health. For a chatbot and a children’s game, those are not abstract categories.
Why these two #
Roblox is an obvious target. It faces lawsuits and criticism over child-safety failures, and has been rolling out age checks and age-separated chat in response. The DSA’s core aim is protecting children online, and Roblox is where millions of them gather.
ChatGPT is the more novel case. OpenAI is fighting wrongful-death suits from families who say the chatbot failed users in crisis, and Florida has sued the company and Sam Altman. Pulling a generative-AI product into the top tier signals that Brussels sees chatbots as platforms to be policed, not tools to wave through.
A familiar fight #
This is now a well-worn pattern. The Commission has opened more than a dozen probes since the DSA took effect. It fined AliExpress €550m, penalised Temu, and hit X with a €120m charge it is appealing. It has escalated an addictive-design probe into Meta and charged TikTok over child safety.
It is also part of a wider European push, from France’s under-15 social-media ban to a growing AI rulebook. The Trump administration calls the DSA a form of “economic extortion” against American firms. Critics ask, more pointedly, why Europe regulates the platforms it cannot seem to build.
The counter is straightforward. As X shows, a designation is a demand, not a cure, and the platform is no one’s model of online safety despite its fine. But forcing two of the internet’s most scrutinised services to open their systems to auditors still beats taking their word that all is well.
Get the TNW newsletter #
Get the most important tech news in your inbox each week.