Google has bought a large cache of data from failed U.S. airline Spirit for $10 million, according to The Register, which cites a court filing tied to the airline’s liquidation process. The report says the data was sold as a deidentified asset after Spirit grounded itself permanently in May 2026 and began auctioning assets to raise cash for creditors. The reported scale is unusually broad for an AI-relevant data purchase. The Register says the trove includes 100 million emails, 500 million Microsoft Teams items, 17 million OneDrive files and 20.5 million SharePoint items. It also includes more than 30 million recorded customer-service calls and more than 15 million customer-service chat records. The dataset is not limited to customer communications. The Register says it also contains 600,000 ServiceNow tickets, 13.7 million active email addresses from Oracle’s Responsys marketing application, and details of 11 million in-flight Wi-Fi sales. Operational records reportedly include data describing more than 763,000 flights, five million crew pairings, more than 1.2 million fuel slips and purchases of 787,452 parts. Google reportedly said it bought the data to improve its AI services. The Register also reports that Mercor, described as a company that provides data to train AI models, was the underbidder. That detail matters because The Register frames the asset as valuable to AI companies. The Register connects the purchase to growing interest in domain-specific AI systems, including smaller models trained on narrower bodies of knowledge. The available summary does not establish what Google will build with the Spirit data, and it would be premature to say the company is making an aviation operations model. What the reported contents do show is a mix of communications, support workflows, sales records and operating data from a real airline environment. Privacy is the obvious risk area. The Register says the court filing describes the dataset as deidentified before sale, and says Google promised to scrub any personally identifiable information it finds in the trove. That leaves the key question unresolved: how complete the deidentification was, and how Google will verify that the data is safe to use in AI systems. For now, the available source material is The Register’s detailed report citing a court document. The hard numbers are specific, but The Register’s report does not establish Google’s precise intended use or the practical privacy controls around the dataset. Who benefits: Google gains access to a large, domain-specific dataset if the sale and reported terms stand as described. Creditors in Spirit’s liquidation may benefit from the $10 million proceeds. Who's exposed: Former Spirit customers may have privacy concerns, although The Register says the data was deidentified and Google promised to scrub PII it finds. The broader privacy risk depends on how complete that deidentification proves to be.