EU labels ChatGPT a VLOSE as new AI Act rules begin

The classification of ChatGPT as a Very Large Search Engine subjects OpenAI to the highest tier of EU digital oversight, requiring proactive mitigation of illegal content and systemic harms. For the streaming and media ecosystem, the Munich court ruling against Suno establishes a critical precedent that training generative models on public platforms like YouTube constitutes copyright infringement if original elements remain recognizable. This tightening of both transparency and intellectual property enforcement signals a shift toward high-friction compliance for AI developers operating in Europe. Watch for the final adoption of the AI Act implementing law in the Czech Republic as a signal for how national authorities will handle local enforcement.

The European Commission's enforcement posture under the Digital Services Act has intensified across multiple platforms in 2025 and 2026. In February 2025, the Commission opened formal proceedings against TikTok under the DSA over addictive design and child safety failures, marking the first time a social media platform faced systemic risk scrutiny under the law. That action followed earlier DSA enforcement against X (formerly Twitter) and Meta, establishing a pattern of escalating oversight that now extends to AI-powered services like ChatGPT. The VLOSE designation framework, originally designed for search engines and social networks, is being applied to generative AI tools for the first time, signaling that the Commission views large-scale AI interfaces as carrying comparable systemic risks to traditional platforms.

The Munich Regional Court's ruling against Suno in August 2025 represents the first major EU judicial decision on generative AI training data and copyright. GEMA, Germany's music rights society, brought the case alleging Suno's AI model had been trained on copyrighted compositions without authorization, and the court found that recognizable musical elements in Suno's outputs constituted infringement. This precedent directly affects streaming platforms and music services that license content, as it establishes that AI companies cannot rely on fair-use-style defenses for training on copyrighted works in the EU. The ruling also intersects with the broader AI Act transparency requirements now in force, which mandate that AI-generated content be clearly labeled, creating a dual compliance burden for generative AI companies operating in European markets.

OpenAI has faced parallel regulatory pressure beyond the DSA. In March 2025, Italy's data protection authority opened an investigation into ChatGPT's data processing practices following a data breach affecting thousands of users, examining whether the company adequately informed users about how their personal data was used for model training. Meanwhile, Reddit and Roblox have both faced scrutiny from EU regulators over AI-generated content moderation failures in 2025, highlighting that the regulatory net extends beyond AI developers to platforms hosting AI outputs. For streaming and media companies, the convergence of DSA enforcement, , and copyright precedent creates a compliance landscape where both AI tool providers and content platforms face simultaneous obligations under multiple regulatory frameworks.