EU Orders Google to Open Android Search Data to AI Rivals
Under DMA enforcement, Brussels is forcing Google to share the query data that feeds AI training — and to stop defaulting Android users toward its own assistant. The mobile ecosystem just became a contested frontier.
The European Union has moved past fining Google and started rewriting its operating conditions. Under Digital Markets Act enforcement, regulators have ordered Google to grant rival AI and search companies greater access to Android search query data — the raw material that trains recommendation systems, assistant models, and next-generation search products. This isn't a consent decree pending appeal. It is an operational mandate aimed at the architecture of Android itself.
What the Order Actually Requires
The instructions target specific data channels on Android — the pathways through which search queries flow and accumulate. Third-party AI assistants and recommendation systems are the named beneficiaries: companies building products that compete directly with Google's own AI features now have a regulatory claim on data they previously couldn't access.
Google is also required to modify default configurations that steer users toward its own AI features. In practice, that means Android must make it meaningfully easier to install and use competing assistants on smartphones — not just technically possible in a buried settings menu, but genuinely accessible. Default placement on a billion-plus device platform is not a minor concession.
The Regulator's Core Argument
EU regulators are making an explicit link between data access and AI capability. Their argument: restricting query data gives Google an unfair structural advantage in developing Gemini-based and other proprietary AI features on Android. The data flywheel — queries generate training signal, training signal improves the assistant, a better assistant captures more queries — is precisely what the order is designed to interrupt.
This framing matters because it moves the enforcement conversation beyond search market share percentages and into the mechanics of AI development. Regulators are not just saying Google is too big in search. They are saying that Google's control of AI infrastructure, data, and distribution through the mobile ecosystem constitutes an unfair advantage in an entirely different product category. That is a materially broader claim.
Why Mobile Is the New Chokepoint
The broader EU push — of which this order is explicitly a part — is focused on preventing dominant platforms from using mobile ecosystems to lock up the next layer of the technology stack. The smartphone is where AI assistants actually reach users at scale. Default assistant status, access to on-device behavioral data, and pre-installation rights are the distribution levers that determine which AI products get traction and which stay marginal.
By targeting Android's default configurations and data architecture simultaneously, Brussels is trying to pry open both the supply side (training data) and the demand side (user access points) at once. Whether that intervention produces genuine competition or merely compliance theater depends entirely on implementation — how the data-sharing protocols are specified, audited, and enforced over time.
The Bigger Shift
What's happening in Brussels is a preview of how regulators globally will eventually engage with AI competition policy. The question is no longer whether a company is dominant in a legacy product category. It is whether control of data pipelines and default placements in existing ecosystems can be used to predetermine outcomes in the AI layer being built on top of them. The EU has decided the answer is yes — and that Android is exhibit one. For every founder or operator building an AI product that touches mobile distribution, the rules of the terrain just changed.