Tech

Flight attendants challenge Google’s $10 million Spirit Airlines data purchase

The Association of Flight Attendants argues that consumer privacy laws are insufficient to protect confidential employee records acquired from the bankrupt carrier.

Editorial persona
Owen Mercer
Markets and Finance Editor
Published
Draft
Source: Ars Technica · View original source
Flight attendants freaked out that Google is buying tons of Spirit employee data
MARKETS & FINANCE

The Association of Flight Attendants (AFA) has lodged a formal court objection to Google’s acquisition of a vast dataset from the bankrupt Spirit Airlines. The union contends that the privacy safeguards embedded in the deal, which rely on consumer protection laws to strip personally identifiable information, fail to adequately protect the confidential nature of employee records.

Google secured the rights to the data at an auction held on 14 August 2026, outbidding rival Mercor Corporation for a sum of $10 million. The dataset comprises approximately 100 million employee emails, human resources files, payroll data, and productivity metrics. While the transaction explicitly excludes consumer data such as customer profiles and loyalty records, it retains decades of workplace communications and operational records.

In its filing, the AFA argued that the "privacy architecture" of the transaction is consumer-facing, yet the "payload" is disproportionately employee-facing. The union stated that de-identification measures, which remove names and direct identifiers, do not address the confidentiality of the content itself. Specific records, including disciplinary correspondence, training deficiencies, and internal communications regarding staffing grievances, remain sensitive even when stripped of individual names.

The AFA is concerned that Google could combine the de-identified worker data with other datasets to infer sensitive details about specific individuals or small groups of flight attendants. The union noted that while Google has committed to never intentionally re-identify the data, the agreement does not prohibit the inference of information about identifiable subgroups. A 2017 study by a Georgia Law researcher highlighted that increasingly powerful hardware makes it easier to de-anonymise scrubbed data when combined with public information.

Google has stated that it acquired the enterprise dataset to improve its artificial intelligence models and other products. A company spokesperson confirmed that the data will be rigorously scrubbed of personally identifiable information by a third party before receipt. Google agreed to use a court-appointed ombudsman to oversee this process and has made binding commitments to maintain the data in a de-identified form.

The Electronic Frontier Foundation has also expressed opposition to the sale, citing the lack of employee consent for the new use of their data. The AFA has requested that the court deny approval of the sale until Google excludes all flight attendant information or extends consumer-level protections to workers. These protections would include prohibitions on using the data to profile, evaluate, or score any individual or identifiable group of former Spirit employees.

Continue reading

More from Tech

Read next: Gardner City Council scraps Flock Safety cameras over privacy concerns
Read next: NASA scraps Swift rescue mission as Link satellite struggles with control issues
Read next: The Binding of Isaac: Repentance+ Online to launch alongside Grand Theft Auto 6