EU Forces Google to Share Search Data with ChatGPT and Claude

The ruling that could break Google's AI monopoly

Think of it like this: imagine you run a restaurant, but the biggest supplier in town refuses to sell you the same ingredients they use in their own restaurant. That's essentially what Google has been doing with its search data and Android features when it comes to AI rivals.

On January 27, 2026, the European Commission said enough. It opened two formal proceedings under the Digital Markets Act (DMA) to force Google to share its search data and open Android features to competitors like ChatGPT, Claude, and Perplexity. The deadline is clear: 6 months to comply or face fines up to 10% of global revenue, which translates to roughly $38.5 billion.

What most guides won't tell you is that this ruling doesn't just affect Google. It could fundamentally change how AI assistants work on your phone and the quality of every Google alternative you use.

What the EU is demanding from Google

1. Open Android to rival AIs

The first proceeding targets Article 6(7) of the DMA and demands something very specific: that third-party AIs get the same access to Android features that Google's own Gemini AI currently enjoys.

The trick is that Google has designed Android to give Gemini advantages its competitors simply can't match:

• System-level API access enabling deep integration
• On-device processing (faster, more private)
• Native integration with calendar, contacts, camera, and other system apps
• System-level voice commands

Think of it like this: what if ChatGPT or Claude could run on your Android exactly like Gemini does—accessing your photos for analysis, reading your calendar to plan your day, or running in the background without opening an app? That's what the EU wants to make possible.

2. Share search data with competitors

The second proceeding is even more explosive. Under Article 6(11) of the DMA, Google must share with competitors:

• Ranking data: how Google ranks search results
• Query data: what users search for (anonymized)
• Click data: which results users click on
• View data: which results users view

Here's the catch: Google claims it already shares data, but what they won't tell you is that their anonymization method is so restrictive it excludes 99% of all queries. Only searches performed more than 30 times in 13 months by more than 30 different signed-in users are included. In other words, only the most generic searches make the cut. The specific, long-tail queries that competitors actually need are left out.

As DuckDuckGo stated in its formal complaint: "Google's dataset would omit a staggering ~99% of search queries including 'long-tail' queries that are the most valuable to competitors."

The timeline: 6 months to change everything

Let me break this down: these are "specification proceedings," not punitive investigations (yet). It's like the EU is saying: "We're going to explain exactly what you need to do. If you don't do it after that, then the fines come." And those fines are anything but symbolic.

The numbers that keep Google up at night

To understand why Google is taking this seriously, look at the data:

Google's market dominance

Potential DMA fines

For context: Google has accumulated over €11 billion in EU antitrust fines over the past decade. But DMA penalties are in a different league entirely. A 10% fine on Alphabet's $385 billion annual revenue (2025 figures) equals $38.5 billion. That's 76 times Apple's DMA fine (€500M) and 192 times Meta's fine (€200M) under the same law.

Who wins if Google complies

The big beneficiaries

OpenAI (ChatGPT): Could integrate ChatGPT at the system level on Android, competing directly with Gemini in features like voice assistant, image analysis, and automation.

Anthropic (Claude): Would gain the same level of Android integration that Gemini currently enjoys, plus search data to improve response quality.

Perplexity: As an AI-powered search engine, accessing Google's ranking and click data would be like finding a treasure map. It could dramatically improve its result quality.

DuckDuckGo, Ecosia, and European search engines: They've been asking for this data for years. With real access (not the current 1%), they could offer far more relevant search results.

What changes for you as a user

If Google complies by July 2026, here's what you could see on your Android:

• Choose your default AI: install ChatGPT, Claude, or Perplexity as your primary assistant, not just Gemini
• Deep integration: your favorite AI accessing calendar, contacts, photos, and system functions
• Better search alternatives: DuckDuckGo, Bing, and others with more relevant results
• More price competition: real competition typically drives down premium subscription costs

Google's response and why it doesn't convince

Clare Kelly, Google's Senior Competition Counsel, responded with a familiar argument: "Android is open by design, and we're already licensing Search data to competitors under the DMA. However, we are concerned that further rules which are often driven by competitor grievances rather than the interest of consumers, will compromise user privacy, security, and innovation."

The trick is to break down this response:

1. "We already share data": Yes, but excluding 99% of queries. That's like saying you share your pizza but only hand over the crumbs.
2. "Privacy": A legitimate argument, but the DMA requires anonymized data. Google is using privacy as a shield to protect its competitive advantage.
3. "Competitor grievances": Competitors are complaining because they can't compete. That's literally the problem the DMA is trying to solve.

As researchers from the SCiDA project noted: "Ignoring AI services would be a fundamental mistake as key competitive forces would fall outside regulatory scrutiny. Without clear requirements, gatekeepers can freely combine massive datasets to train superior models while competitors cannot access equivalent data."

The bigger picture: EU vs Big Tech in 2026

This action against Google doesn't exist in a vacuum. The EU has intensified its offensive against major tech companies:

The Commission describes 2026 as the year "enforcement becomes the priority." In May 2026, the first comprehensive DMA review milestone arrives, and there are clear signals that regulatory intensity will increase, not decrease.

If you're concerned about how Google handles your privacy in general, we've already covered Gemini reading your emails without permission and the Gemini Calendar vulnerability.

US vs Europe: two opposing philosophies

This is where things get geopolitically interesting.

Europe (proactive regulation): Sets rules before harm occurs. Identifies gatekeepers and imposes immediate obligations. The philosophy: "markets need guardrails to be fair."

United States (reactive enforcement): Waits for demonstrable harm, then litigates case by case. Antitrust proceedings can take 5-10 years. The philosophy: "markets self-correct."

The tension between these approaches has intensified in 2026. The Trump administration views European digital regulations as "economic warfare" and "non-tariff barriers," threatening retaliation and tariffs against the EU for its actions against Silicon Valley.

The EU, for its part, insists its regulations promote safety, fair competition, and consumer rights—not protectionism. The outcome of this standoff could define the future of global tech regulation.

What the DMA means for the future of AI

Optimistic scenario

If the DMA works as the EU intends:

• Real competition: ChatGPT, Claude, and Perplexity compete on equal footing with Gemini on Android
• Better products: alternative search engines improve dramatically with access to real data
• Diverse innovation: specialized AIs (privacy, health, education) can flourish
• Lower prices: competition drives down premium subscription costs

Pessimistic scenario

If the DMA generates unintended consequences:

• Privacy risks: more companies handling user data (even anonymized) increases the attack surface
• Fragmentation: too many AI options could confuse the average user
• Investment disincentive: if Google can't capitalize on its data advantage, it might reduce investment in Android and Search
• Trade war: EU-US tensions could spill into other sectors

My analysis

Let me break this down: the DMA is imperfect but necessary. Google controls 90% of searches and 72% of mobile devices. Without regulation, generative AI will become another de facto monopoly, because whoever has the data trains the best models, and whoever has the best models attracts more users, who generate more data. It's a cycle that only regulation can break.

Will the DMA solve everything? No. Google has infrastructure advantages (TPUs, data centers, talent) that no law can level. But opening access to data and system features is a critical first step toward real AI competition.

If you're interested in how AI is reshaping e-commerce, check out how Google plans to use AI agents for shopping. And if you're an Android user, you might want to know that Chrome can already browse for you with Gemini.

Frequently Asked Questions (FAQs)

What is the Digital Markets Act (DMA)?

The DMA is a European Union regulation in effect since May 2023 that imposes obligations on major tech platforms designated as "gatekeepers" (Google, Apple, Meta, Amazon, Microsoft, ByteDance, and Booking.com). Its goal is to ensure fair and open digital markets by preventing these companies from abusing their dominant position.

When does Google have to comply with the EU's demands?

The European Commission opened proceedings on January 27, 2026, with a 6-month deadline. Google should have everything implemented by approximately July 2026. Non-compliance would trigger a formal investigation that could result in fines up to 10% of global revenue (approximately $38.5 billion).

Does this mean ChatGPT will work like Gemini on Android?

If Google complies, yes. The EU requires that third-party AIs have "equally effective access" to the Android features that Gemini uses. This would include system-level integration, on-device processing, and access to features like calendar, contacts, and camera.

Will my search data be shared with other companies?

The data must be shared in anonymized form under "fair, reasonable, and non-discriminatory" (FRAND) terms. No personally identifiable data will be shared. The current debate is about how much anonymization is enough: Google currently excludes 99% of queries with its method, and the EU considers this insufficient.

What other companies have been fined under the DMA?

Apple was fined €500 million in April 2025 for anti-steering restrictions in the App Store. Meta received a €200 million fine that same month for its "pay or consent" model. Google has preliminary self-preferencing findings in search (fine pending) and now faces these new proceedings on Android and AI.