Google Buying Spirit Employee Data Sparks Flight Attendant Revolt
Flight attendants object to Google's $10M purchase of Spirit Airlines worker data, citing privacy loopholes. They ask court to block sale until protections are extended.
Google Buying Spirit Employee Data Sparks Flight Attendant Backlash
Google buying Spirit employee data has triggered an urgent legal objection from the union representing the airline’s former flight attendants. Last Friday, the tech giant won a bankruptcy auction to acquire a massive dataset covering nearly the entire employment and workplace record of the defunct carrier. But the deal explicitly excludes customer information and personal data. That’s a key caveat. Google agreed to a court-appointed ombudsman who will oversee stripping any personally identifying information before transfer, a process that could take months, given the volume of records involved, and it’s unclear how the ombudsman will verify compliance. So the union’s challenge hinges on privacy safeguards. It can’t trust the plan.
The company also pledged to keep the data de-identified and never intentionally re-identify it. Third parties buying access would face the same restrictions. On paper, that sounds airtight. But the Association of Flight Attendants (AFA) sees a glaring hole. Consumer privacy laws protect passengers. Worker confidentiality appears to get no such shield.
The Privacy Loophole at the Heart of the Deal
In a court filing Tuesday, the AFA argued that Google leaned on consumer protection statutes to guarantee the dataset contained no PII. Those laws, however, do not cover employee records. The union’s filing put it bluntly: the privacy architecture is consumer-facing while the payload is disproportionately employee-facing.
“Hence, the employee data is far more confidential than the customer data, yet receives far less protection than the customer data.”
The math is stark.
Adam Schwartz, a privacy litigation director at the Electronic Frontier Foundation, told Ars that the sale alarmed privacy advocates, and he opposed using a person's data for a new purpose without consent, especially when a bankrupt company sells employee emails to become AI training data. It's a betrayal. But he didn't mince words. That sort of repurposing shouldn't happen, full stop.
How Google Won the Auction
The virtual auction took place August 14. Google opened with a $5 million bid, promising from the start to cover the cost of a third party scrubbing the data. Competing bids were rejected after requesting additional consumer data, including a specific customer list Spirit refused to sell. Any bid seeking PII got cast aside after the first round.
Google spent two and a half hours fighting off rivals. Mercor Corporation proved the fiercest challenger, repeatedly offering to scrub the data itself. That self-scrubbing proposal was rejected each time. Google ultimately won with a $10 million price tag, plus the extra step of hiring an independent service to comply with consumer privacy laws. An alternative bid with similar terms for $7.5 million was accepted from Mercor in case Google fails to follow through.
Employee Data vs. Consumer Data
The AFA's objection is limited. It doesn't seek to disrupt the sale entirely. But the union argues that Spirit's debtors should have protected workers just as strongly as they protected consumers, wrapping employee safeguards into the deal's fabric rather than leaving them as an afterthought. So the sale agreement nowhere requires anyone to screen for, segregate, or restrict the use of confidential employee information,and that's a glaring hole in the plan.
De-identification only tells you if a record can be traced to a named individual. It doesn't touch the real problem. But the contents can stay confidential regardless. A flight attendant's disciplinary correspondence, a crew training deficiency, a payroll adjustment history , each remains sensitive even when the employee's name is missing, and that sensitivity doesn't vanish just because the identifier is stripped away. So we can't pretend the risk is gone. It isn't.
Workers worry Google could merge this data with other datasets it already holds. But re-association might happen passively, despite the company's promise to never intentionally re-identify anyone. It's a real risk. A Georgia Law researcher noted back in 2017 that increasingly powerful computer hardware makes it easier to combine publicly available data with scrubbed records to de-anonymize them, and that's a process we can't fully control. So the threat isn't just deliberate. It's accidental. And that makes it harder to stop.
What Flight Attendants Fear Most
The union does not make a technical claim that any particular record can be re-identified. But they argue the risk is not speculative. Where a small, highly structured population is described across linked operational and communications datasets spanning more than a decade, inference becomes a real threat.
Google’s covenant reaches only intentional association. That overlooks the possibility that confidential data could expose information about identifiable groups. The AFA wrote that a pseudonymized dataset could still disclose which crew bases generated grievances, how a small subset of flight attendants performed on recurrent training, which employees were subject to investigation, and what compensation adjustments followed which events.
Even without names, sensitive information that has nothing to do with names still passes through untouched. That's the consequence. But the union warned that only Google has a voice in how de-identification works, a process that ignores legitimate concerns of parties who aren't at the negotiating table, and the court should acknowledge that as a problem. It's a glaring oversight. So Google's unilateral control shapes the outcome, and that's not fair to anyone else involved.
Google’s Response and the Trust Problem
A Google spokesperson said the company acquired part of an enterprise dataset from Spirit Airlines to improve its products and AI models. The company will not receive any personal information. Any data received will be rigorously scrubbed of PII by a third party before receipt.
The data won't reach Google until it's scrubbed. So the original identifiers are off-limits, permanently, unless a court orders otherwise. A source close to the sale, granted anonymity to discuss technical risk, said Google’s vendors will apply certified de-identification industry standards, and the company also made binding commitments in court to never intentionally re-identify the data. That's a hard promise.
But skeptics remember Google’s history. In 2024, the company settled a class action lawsuit raised by Incognito users, agreeing to delete billions of data records reflecting private browsing activities. Last year, Google agreed to pay Texas $1.4 billion to settle claims it unlawfully tracked geolocation, incognito searches, and biometric data. Texas Attorney General Ken Paxton called it the highest recovery nationwide against Google for any attorney general’s enforcement of state privacy laws.
The AFA’s Demands and What Comes Next
The union asked the court to deny approval of the sale until Google agrees to exclude all flight attendant information from the purchase. They also want notification to workers when Google allows a third party to access the data. It's unclear if the court will weigh in. But the filing came one day after the court’s deadline to object, and given that the deadline has already passed, the judge now faces the question of whether to accept this late submission or simply let the sale proceed without further review.
A hearing is scheduled in September where the court will possibly approve the sale. The AFA suggested a representative would be present. Their filing argued that if the court permits the sale, it should only do so with the same protections extended to consumers extended to former Spirit flight attendants.
The order must directly prohibit Google from using worker data to analyze, profile, evaluate, score, or draw conclusions about any individual Spirit flight attendant. That's the core demand. It should also bar any attempts to re-associate that de-identified data with a specific Spirit employee, closing every loophole that might allow the company to quietly piece identities back together later. So don't let them wiggle out. Keep it airtight.
The flight attendants’ core argument cuts to the heart of the transaction. Nearly every consumer-facing category is designated “Not Included.” Customer Profiles, loyalty and Free Spirit data, active email addresses, chat sessions, call recordings, telephone numbers, website analytics, DOT complaints. All excluded.
Nearly every category under “Team Member” is designated “Included.” Time card information, employee data and records, business travel records, corporate and crew training records, payroll records, employee tax forms, and employee documents. All included.
That asymmetry is why Google buying Spirit employee data raises such alarm. The deal asks both consumers and workers to place enormous trust in a company with a checkered privacy past. And for the people who actually worked at Spirit, the stakes feel existential. Their confidential records, their disciplinary histories, their private communications. All now sitting in a dataset sold to the highest bidder. Names removed, but sensitivity intact. The court will soon decide whether that counts as protection or merely a technicality.
The union's challenge hinges on privacy safeguards because it cannot trust the plan to strip personally identifying information from the data. The article states the deal explicitly excludes customer information but includes employee records, and the union sees a glaring hole in the protection of worker confidentiality compared to consumer data. Google won the auction by opening with a $5 million bid and promising to cover the cost of a third party scrubbing the data. After two and a half hours of fighting off rivals, Google ultimately won with a $10 million price tag and the extra step of hiring an independent service to comply with consumer privacy laws. The AFA argues that the privacy architecture is consumer-facing while the payload is disproportionately employee-facing. In their filing, they stated that employee data is far more confidential than customer data, yet receives far less protection, noting that disciplinary letters or training deficiencies remain sensitive even when names are stripped. Flight attendants fear that Google could merge this data with other datasets it holds, and re-association might happen passively despite the promise to never intentionally re-identify anyone. The article notes that increasingly powerful computer hardware makes it easier to combine publicly available data with scrubbed records to de-anonymize them, and the union argues the risk is not speculative because inference becomes a real threat in a small, structured population. The AFA asked the court to deny approval of the sale until Google agrees to exclude all flight attendant information from the purchase. They also want the order to directly prohibit Google from using worker data to analyze, profile, evaluate, score, or draw conclusions about any individual Spirit flight attendant, and to bar any attempts to re-associate de-identified data with a specific employee.Frequently Asked Questions
What triggered the union's legal objection to Google buying Spirit employee data?
How did Google win the bankruptcy auction for Spirit's employee data?
Why does the AFA argue that worker data requires more protection than consumer data in this deal?
What specific risk do flight attendants fear regarding the de-identified data, according to the article?
What are the AFA's core demands in their court filing?
Valerie Dubois covers public policy and regulation, with a focus on how decisions made by governments affect technology and society. She follows the debates that shape the rules we all live by.
💬 Comments (0)
No comments yet. Be the first!
