EU Investigates Google Over AI Content Use and Competition Rules

Why the EU Investigates Google Amid Rising Concerns Over AI Data Use

Brussels launches another major probe into how Google gathers and uses online content to train its rapidly expanding AI systems. The move marks a sharp escalation in the battle between regulators and Big Tech, especially as advanced computing models like Gemini, AI-Overviews, and generative summaries reshape the information ecosystem.

Source: X

But, why is the EU investigates Google again, and what does this latest inquiry mean for publishers, creators, and the future of Artificial intelligence in Europe?

EU Investigates Google Amid Rising Concerns Over AI Data Use

The European Commission has opened a new antitrust probe into whether the Alphabet's division used publishers’ content and YouTube videos to train its AI models without proper consent or compensation. Regulators are particularly focused on company’s intelligence system overviews, which generates summaries above search results in more than 100 countries.

The EU investigates Google not only over data scraping but also over how AI-generated content is labeled. Officials fear Artificial intelligence summaries could be mistaken for real journalism, hurting transparency and diverting revenue from publishers.

Key concerns include whether:

• Google gives itself preferential access to training data

• Rival developers face unfair disadvantages

• YouTube creators have their content used for smart algorithms without consent

If the tech giant is found guilty, it could face fines up to 10% of its global annual revenue, one of the heaviest penalties in European Union competition law.

The competition law chief Teresa Ribera said the Commission fears the platform may be “abusing its dominant position” by imposing unfair conditions on publishers, allowing the platform’s AI-powered tools to benefit from content creators without ensuring fair economic return.

Google rejected these accusations, arguing the complaint risks “stifling innovation” and insisting Europeans should not be denied access to the latest Next-gen intelligence technologies.

On the other hand, groups like the Independent Publishers Alliance and Foxglove supported the EU’s worries and argued the searching giant is using website content without proper licensing with: 

• AI Overviews reduce traffic to original sources

• Publishers lose revenue while the platform monetizes AI-outputs

• YouTube creators have no clarity on how their work is used in training datasets

One lawyer labeled Gemini as “Search’s evil twin,” claiming it drains value from creators while replacing their content with artificial intelligence summaries.

A New Front in the ML-Driven Systems Regulatory War

This case comes just as the EU prepares to enforce its landmark AI Act, requiring transparency around training data, model risk assessments, and lawful data use. Combined with GDPR, Europe now holds the world’s strictest AI compliance regime.

The investigation also deepens ongoing tension between Brussels and Washington.
In recent months:

• The social media platform, X, was fined €120M

• Meta came under scrutiny for allowing machine intelligence providers access to WhatsApp

• Google, before this, was already fined nearly €3B for ad-tech dominance
  

In response to the probes, the U.S. lawmakers accused the EU of targeting American tech giants. 

Why This Investigation Matters for AI’s Future

The EU investigates Google at a moment when global competition is intensifying and nations are racing to control how training data is harvested.

The probe signals a major turning point: governments now challenge not just what AI-systems generate, but what data they are trained on.

This shift could influence:

• How AI-powered models are built

• Whether scraping public content remains legal

• How creators are compensated

• How Intelligent system's transparency is defined by law

If the EU investigates Google successfully and imposes penalties or obligations, it could force every major AI company – OpenAI, Meta, xAI, Anthropic, to overhaul their training pipelines.