- The European Commission is preparing to designate OpenAI’s ChatGPT and Roblox as “very large online platforms” (VLOPs) under the Digital Services Act, a designation that triggers the EU’s most stringent tier of content moderation and transparency obligations and is expected to be formalized as soon as August; the VLOP threshold is 45 million monthly active users in the EU, and both ChatGPT and Roblox have now crossed it — ChatGPT’s crossing reflects the rapid mainstream adoption of AI assistants across European consumers and enterprises since its 2022 launch, while Roblox’s crossing reflects its dominant position in the youth gaming market across Europe; once designated, both companies will join Meta, X, Google, TikTok, and other existing VLOPs in facing the full weight of DSA compliance: mandatory transparency reports, detailed algorithmic risk assessments, risk mitigation plans filed with the Commission, annual supervisory fees paid to the European Commission, and potential fines of up to 6% of global annual revenues for violations.
- The ChatGPT designation is particularly significant because it applies the DSA’s platform regulation framework — originally designed around social networks and search engines — to a generative AI system, creating novel compliance questions that the DSA’s original drafters did not specifically contemplate; the DSA’s content moderation obligations (clamp down on illegal and harmful content, provide algorithmic transparency, allow users to opt out of recommendation systems) were written with social media feeds and search results in mind, and applying them to a conversational AI system that generates content rather than distributing user-generated content requires regulatory interpretation that neither OpenAI nor the Commission has fully worked through; the designation will force OpenAI to engage with EU regulators on questions including how to provide “algorithmic transparency” for a large language model, what constitutes an adequate risk mitigation plan for AI-generated content, and how DSA’s child protection requirements interact with ChatGPT’s existing age verification and content policies.
- The Roblox designation is less regulatory novelty but carries significant child safety implications: Roblox has faced sustained criticism over its child safety protections, including a large volume of child sexual abuse material (CSAM) cases that the company has been attempting to move out of court proceedings, and has recently implemented stronger controls over who children can communicate with and what games they can access; the DSA’s child protection requirements — which compel VLOPs to safeguard minors’ experiences and restrict targeted advertising to children — will apply with full force once the designation is formalized, and Roblox’s recent moves into advertising (it overhauled its ad policies in early 2026 to boost sponsorship revenue) will face specific DSA constraints on marketing to the platform’s predominantly young user base; the Commission’s designation signals that Roblox’s child safety improvements have not been sufficient to avoid the strictest regulatory tier.
- The Trump administration’s reaction to DSA designations has been consistently hostile: the administration has characterized DSA enforcement against American companies as EU censorship of U.S. tech and has raised the issue in trade negotiations; with ChatGPT — a flagship American AI product — now being pulled into the DSA’s strictest compliance tier alongside Meta, X (already fined €120 million for deceptive design), and Alibaba (fined €550 million in July for counterfeit product failures), the diplomatic friction between Washington and Brussels over digital regulation is likely to intensify; the 6% of global revenue fine ceiling for DSA violations creates a material financial risk exposure for OpenAI specifically, given its rapidly growing but not-yet-disclosed revenue base and its ongoing transition from nonprofit to for-profit structure.
What Happened?
The European Commission will designate OpenAI’s ChatGPT and Roblox as “very large online platforms” under the Digital Services Act as soon as August, after both platforms surpassed 45 million monthly EU users. The VLOP designation triggers the DSA’s strictest compliance tier: transparency reports, algorithmic risk assessments, risk mitigation plans, annual supervisory fees, and potential fines of up to 6% of global annual revenues for violations. The commission has already fined X €120 million and Alibaba €550 million under DSA in 2026.
Why It Matters?
The ChatGPT designation is a regulatory landmark: it applies a platform content moderation framework originally designed for social media to a generative AI system, creating entirely new compliance territory around algorithmic transparency, AI-generated content risk assessment, and child protection obligations for conversational AI. For Roblox, it intensifies scrutiny on child safety practices and advertising restrictions at a moment the company is expanding into ad revenue. Both designations add to the Trump administration’s diplomatic friction with Brussels over DSA enforcement, which Washington has characterized as targeted censorship of American tech companies.
What’s Next?
Watch for the formal August designation announcement from the European Commission and OpenAI’s/Roblox’s public responses — any hint of legal challenge to the designation would escalate the U.S.-EU digital trade dispute; watch for the Commission’s specific guidance on how DSA transparency and risk mitigation obligations will be interpreted for generative AI systems, which will set the precedent for how Anthropic, Google Gemini, and other AI assistants above the VLOP threshold are regulated; watch Roblox’s advertising revenue disclosures for any indication that DSA child advertising restrictions are materially affecting its sponsorship revenue strategy; and watch the Trump administration’s trade negotiation posture toward the EU, where DSA enforcement against American AI companies may become a more prominent bargaining chip.
Source: Bloomberg
