The European Commission has issued a formal notice to major technology companies, stating that the integration of artificial intelligence features into established services like search engines and social media platforms is being closely monitored for compliance with the European Union's digital competition laws.
In a letter to members of the European Parliament, the EU's tech commissioner confirmed that existing rules under the Digital Markets Act (DMA) apply directly to how these companies deploy AI, particularly concerning data use and user choice.
Key Takeaways
- The European Commission is actively monitoring how Big Tech companies like Google and Meta integrate AI into their core services.
- EU officials state that these AI integrations must comply with the existing Digital Markets Act (DMA).
- Key concerns include how companies process and combine user data for AI training and operation.
- The DMA requires that users must be able to change or remove default services, including embedded AI features.
EU Officials Confirm Scrutiny of AI Features
The European Commission is paying close attention to the rapid deployment of artificial intelligence by large technology firms. In a letter dated September 24, tech commissioner Henna Virkkunen informed European Parliament members that the executive body is ensuring that AI development and integration adhere to EU law.
Virkkunen's communication specifically addresses the trend of embedding AI into widely used platforms. She noted, "We observe from market developments that in a number of instances AI is being integrated into other services, such as online search engines or online social networks."
This directly relates to recent product updates from major industry players. For example, Google has started placing AI-generated overviews at the top of its search results, while Meta has integrated its AI Assistant across its family of apps, including Instagram and WhatsApp.
Background: The Digital Markets Act (DMA)
The Digital Markets Act is a landmark piece of EU legislation designed to ensure fair and contestable digital markets. It designates certain large online platforms as "gatekeepers" and imposes a set of specific obligations and prohibitions on them to prevent anti-competitive behavior. Companies like Google, Meta, and Microsoft are among those designated as gatekeepers.
DMA Obligations Applied to Artificial Intelligence
The letter from Commissioner Virkkunen serves as a reminder that several Big Tech companies leading AI development are already classified as dominant "gatekeepers" under the DMA. This designation means they are subject to a strict set of rules governing their operations within the EU.
Virkkunen emphasized that these existing obligations are fully applicable to new AI features. The rules are particularly relevant to how these companies handle vast amounts of user data, which is essential for training and running advanced AI models.
Data Handling and User Consent
A central pillar of the DMA is its regulation of data. The act places strict limits on how gatekeepers can process, combine, and cross-use personal data collected from their different services. For example, data from a social media service cannot be automatically combined with data from a search service without explicit user consent.
These rules now extend to AI. If a company uses data from multiple services to train a single, integrated AI assistant, it must comply with the DMA's data-handling requirements. This is intended to prevent dominant firms from leveraging their vast data ecosystems to gain an unfair advantage in the emerging AI market.
Fact: Review of Competition Rules
In July, the European Commission initiated a formal review of its digital competition framework. A primary focus of this review is to assess how current regulations can foster a "contestable and fair AI sector in the EU," signaling a proactive approach to governing the new technology.
User Choice and Default AI Services
Another critical obligation highlighted in the commissioner's letter concerns user choice. According to Virkkunen, the DMA requires that platforms must allow users to easily change default services. She argued that this principle extends to artificial intelligence features that are embedded within an application or operating system.
This implies that users should have the ability to switch to a different AI assistant or disable the default one provided by the platform. This provision is designed to prevent gatekeepers from locking users into their own AI ecosystem and stifling competition from smaller AI developers or alternative services.
"Big Tech is now trying to extend its dominant position to the AI market by stealing our data and using it to train their AI or integrate it into our apps without our permission. The Commission is clear: This is not allowed under the DMA."
Political Support for Regulatory Oversight
The Commission's proactive stance has received support from lawmakers who have been advocating for stronger oversight of AI. Kim van Sparrentak, a Dutch Greens member of the European Parliament, praised the Commission's clarification.
Van Sparrentak, who had previously urged the Commission in February to explicitly bring AI under the scope of the DMA, stated that the move was necessary to prevent market domination. In a statement, she expressed concern that large firms were using their existing market power to control the future of AI.
The Commission's letter, combined with its ongoing review of competition rules, indicates that the EU is committed to applying its existing legal frameworks to the challenges posed by artificial intelligence. This approach aims to regulate the impact of AI on market competition without waiting for new, AI-specific legislation to be passed and implemented, ensuring that rules keep pace with technological advancement.