Google's $10M Spirit Airlines Data Deal Ignites Labor Fight Over Employee Privacy

Google paid $10 million this summer to acquire more than three decades of Spirit Airlines' operational records in bankruptcy proceedings—a transaction that includes far more than flight logs and invoices. The dataset contains over 1 million time-card records, 175,000+ employee files, 80,000 email accounts, and hundreds of millions of Microsoft Teams and SharePoint files. A Google spokesperson framed the purchase as raw material for improving the company's products and AI systems, and stated that no customer personal information would be included.

The union representing Spirit's 5,500 former flight attendants has other concerns. Days after the court approved Google's $10 million bid over a competing $7.5 million offer from AI data firm Mercor, the Association of Flight Attendants lodged a legal objection, arguing that employee data shouldn't be fair game for AI training—especially not without explicit consent from workers whose careers have ended.

What the Data Actually Contains#

According to court filings, the employee information slated for AI training encompasses sensitive personal details that workers shared with the airline under employment and insurance agreements. Flight attendants say their email and file accounts hold records of health conditions, miscarriages, domestic violence disclosures, and internal union strategy discussions. One anonymous former attendant told the publication, "It never crossed my mind that they would be so bold as to sell our private data for AI."

The union's legal challenge argues that even if Google promises to "deidentify" the data—stripping out names and direct identifiers—the risk of re-linking private information remains high. Modern AI systems excel at connecting disparate data across multiple datasets, they note, making anonymization promises insufficient.

A Broader Gap in Labor Protections#

The Spirit case exposes what legal scholars describe as a stark imbalance in data protection law. Consumer information gets regulatory scrutiny when companies fold or change hands; worker data has largely fallen through the cracks. Seema Patel, a law professor at UC San Francisco who focuses on labor and technology, observed, "There is no boundary between the information and data that the employee is producing and their own personal information. The law has not caught up."

The confrontation marks what experts say is the first major labor union pushback against corporate AI data harvesting, coming as startups race to monetize old Slack archives, GitHub repositories, and defunct company servers. Meanwhile, other tech firms are actively gathering video of workers performing physical tasks—cooking, cleaning, factory work—to train robots that might eventually displace them.