Google Asks EU Court to Suspend Order Forcing It to Share Search Data With AI Rivals

Google has formally petitioned the European Union’s General Court in Luxembourg to suspend a regulatory order that would force the company to share its vast trove of search data with rival artificial intelligence developers. The request for interim relief, filed on September 29, 2026, seeks to pause the mandate while Google’s broader legal challenge against the European Commission proceeds.[1][5]

Under Article 278 of the Treaty on the Functioning of the European Union, filing an appeal does not automatically suspend a Commission decision. Unless the General Court grants the temporary injunction, Google remains legally bound to comply. The company must finalize its anonymized search dataset by November 2026 and begin sharing it by January 2027.[5][6]

The July Mandate

The underlying dispute stems from two binding decisions issued by the European Commission on July 16, 2026, under the Digital Markets Act. The DMA serves as the bloc’s regulatory framework designed to limit the market power of designated technology platforms. The Commission argues that Google’s scale makes it impossible for smaller AI startups to train competitive models without historical data.[1][5][6]

The first directive requires Google to provide eligible competitors—specifically including AI chatbots that offer online search functions—with anonymized ranking, query, click, and view data. The data must be offered on fair, reasonable, and non-discriminatory terms. Regulators stipulate that recipients cannot use the dataset to train general-purpose AI models from scratch, but rather to optimize their specific retrieval technologies.[5][6]

A second order mandates that Google grant third-party AI assistants the same deep integration into its mobile operating system currently enjoyed by its own Gemini models. By August 2027, rival developers must be given access to 11 core Android features. These include voice activation and the ability to execute actions within other applications.[5]

The Privacy Defense

Google contends that complying with the data-sharing requirements before the court rules would expose millions of European users to significant security vulnerabilities. The company argues that the sheer volume and specificity of search histories make true anonymization nearly impossible. Google warns that sophisticated AI models could potentially de-anonymize the data by cross-referencing rare search terms.[1][3]

Oliver Bethell, Google’s senior director of competition, emphasized the sensitive nature of the information users type into the search bar. People use the platform for their most personal questions, ranging from medical worries to close relationships, he noted.[5]

"Mandating we share these personal queries without adequate safeguards would cause irreversible harm to user privacy," Bethell said in a statement accompanying the filing. He added that the secondary order regarding Android integration would equally weaken vital security protections on the mobile operating system.[5]

The European Commission firmly disputes Google’s assessment, arguing that the mandated anonymization framework was developed in direct consultation with independent privacy experts. Under the EU rules, Google is required to strip out usernames, IP addresses, and direct identifiers before any data changes hands. Records containing unusually long queries or sensitive information like bank details must also be entirely suppressed.[5][6]

Industry Divides

The Commission maintains that the remaining data cannot be readily connected to individuals, and that Google retains the right to refuse data access to any specific company it deems a serious security threat. Privacy-focused search engine DuckDuckGo, which does not track its users, has publicly backed the regulator’s position.[5]

A DuckDuckGo spokesperson stated that the anonymization rules are sound and characterized Google’s privacy criticism as unfounded. The explicit inclusion of AI chatbots in the DMA order highlights how regulators view the future of online information retrieval. Traditional search engines return a list of links, but generative AI assistants synthesize answers directly.[3][5]

By forcing Google to share its underlying ranking data with these AI rivals, the European Union is actively attempting to accelerate the transition toward conversational search interfaces. The data sharing mandate specifically targets the long tail of search queries, which are crucial for training AI models to handle nuanced, conversational prompts.[3][6]

While competing search engines might have enough data to accurately answer common queries like weather or sports scores, they lack the volume required to understand complex, multi-part questions. The Commission believes that granting access to Google's click-and-view metrics will bridge this gap, allowing smaller European AI startups to compete against established tech giants.[3][5]

A Global Regulatory Push

Google has countered that forcing it to subsidize its competitors by handing over proprietary data undermines the incentive to innovate. The company argues that it invested billions of dollars and decades of engineering effort to build its search index and ranking algorithms. From Google's perspective, the DMA order effectively penalizes its success.[1][4]

The pressure on Google extends beyond the immediate logistical challenges of anonymizing its data. The European Commission opened specification proceedings against Google in January 2026 under Article 6(11) of the DMA. The provision explicitly requires designated gatekeepers operating online search engines to provide third-party search providers with access to data on fair, reasonable, and non-discriminatory terms.[6]

Regulators argue that no smaller rival could ever build up search data on Google’s scale organically. Giving competitors access to this historical data will mean more choices for people picking between search engines and AI assistants. The Commission insists that its measures will increase competition while keeping privacy and security protections firmly in place.[5]

The General Court in Luxembourg has not yet announced a timetable for ruling on the interim relief request or the broader legal challenges. However, decisions on temporary injunctions are typically expedited. If the court denies the request, Google will be forced to proceed with the complex technical preparations required to open its data vaults.[1][5]