According to Silicon Republic, the European Commission launched a formal antitrust investigation into Google on December 9th, focusing on its use of online content to benefit its artificial intelligence services. The probe specifically examines Google’s use of web publisher content for features like AI Overview in search, and its use of YouTube videos to train generative AI models like Gemini, allegedly without providing appropriate compensation or an opt-out. The EU is concerned Google imposes unfair terms on creators while giving itself “privileged access” to this data, which could disadvantage rival AI developers who are barred by YouTube’s policies from using the same content. This investigation follows a €2.95 billion fine issued to Google in September for adtech antitrust breaches and another probe launched last month into Google allegedly demoting news sites in search. There’s no legal deadline for the investigation’s conclusion, as its duration will depend on the case’s complexity and Google’s cooperation.
The Core Issue is Access and Control
Here’s the thing: this isn’t just about copyright. It’s about market power. The EU’s argument hinges on a classic antitrust theory—tying. Google owns the platform (YouTube) and sets the rules. Users grant Google a license to use their uploaded content for things like AI training. But that same license isn’t available to competitors. So Google gets a massive, perpetually refreshed dataset for free, while everyone else is locked out. That creates a huge moat. If you’re a startup trying to build a multimodal AI that understands video, where do you get your training data? Not from the world’s largest video repository, that’s for sure. The EU seems to be asking: is Google’s dominance in video hosting being unfairly leveraged to create dominance in AI?
A Recurring Theme with Google and the EU
Let’s be honest, this feels like a sequel. The EU and Google have been in this dance for over a decade, with massive fines over shopping comparisons, Android, and now adtech. The pattern is clear: the Commission identifies a market where Google is dominant, then alleges it uses that dominance to unfairly advantage its services in another market. This AI probe is just the latest chapter. And the political rhetoric is heating up, with EVP Teresa Ribera tying “a free and democratic society” directly to this investigation. That’s heavy stuff. It signals they’re not messing around and are framing this as a fundamental issue for Europe’s digital sovereignty, not just a dry competition complaint.
What This Means for Creators and the Web
The big, unanswered question is: what does “appropriate compensation” even look like? For a publisher whose article is scraped, maybe there’s a model. But for a YouTube creator with a single video in a dataset of billions? The logistics of micro-royalties are a nightmare. I think the more likely outcome, if the EU prevails, is a forced licensing regime. Google might have to offer a way for competitors to also license YouTube content for AI training, probably for a fee. That could create a new revenue stream, but it’s messy. And what about the opt-out? If every major publisher and creator opts out, does the quality of AI models plummet? This probe could force a massive rethink of the entire “scrape now, ask later” model that has fueled the AI boom.
A Long Road Ahead
Don’t expect a quick resolution. Antitrust cases like this move at a glacial pace. Google will fight this vigorously, arguing that its use of data is covered by existing terms of service and that it’s investing heavily in AI tools for creators. But the pressure is undeniably building. Between this, the adtech fine, and the Digital Markets Act forcing interoperability, Google’s cozy relationship with the European market is getting a serious overhaul. The real impact might be felt long before a final ruling, as the very threat of this probe could push Google and other giants to start cutting voluntary deals with publishers and creators—just to get ahead of the regulatory curve.
