Data has often been called the new oil, but Google’s $10 million bid for a bankrupt airline’s digital archives is an eye-opening example of just how much old corporate information could be valuable in the era of artificial intelligence.
Spirit Airlines is going through bankruptcy and its assets are drawing attention beyond aircraft, airport slots and other conventional airline assets. And one of the assets that is said to be of interest is a vast archive of corporate data that has accrued over decades. Google offered $10 million for access to emails, Microsoft Teams conversations, employee records, software code and other operational data, among others.
The reported deal is part of a trend in the technology industry: AI companies are increasingly looking beyond publicly available internet content for material that could potentially help develop, evaluate or improve artificial intelligence systems.
What Could Google Get for $10 Million?
The reported package is huge. It is said to contain about **100 million company emails, 500 million Microsoft Teams messages, 30 million lines of code and records with more than 175,000 employees**.
The collection could provide a detailed historical record of how a major airline operated, communicated and developed its technology systems. And unlike internet data, corporate archives can contain information about workflows, customer service interactions, software development, business decisions and operational processes.
The size of the information is particularly significant because Spirit's digital history spans many years. Such archives could provide valuable material for understanding real-world business operations, though the exact way Google intends to use the data has not been publicly identified.
A Bidding War for Corporate Data
The transaction has turned into a rather unusual auction. Google initially bid about $5 million, according to the reports. AI data company Mercor then entered the contest with a much better offer of around $5.2 million.
The bidding continued, with an offer reaching **$7 million** before Google raised its offer to $10 million.
That may seem small compared to the huge amounts technology firms spend on infrastructure and AI development but it is a significant valuation for data from a bankrupt company.
The development also shows bankruptcy proceedings can expose digital assets that may have been overlooked in earlier times. In the past, corporate archives were primarily used for legal, accounting or historical purposes. But in the AI economy, massive collections of structured and unstructured information can be of a completely different value.
Why AI Companies Want More Data
The AI industry's appetite for data has intensified as companies develop increasingly sophisticated models. Large amounts of the easily accessible information on the public internet have already been collected, analyzed or incorporated into existing datasets.
That has led tech companies and AI-oriented businesses to look for alternative sources. Proprietary databases, video footage, specialist documents, human-generated records and other forms of real-world information can provide new material for training and testing AI systems.
The Spirit archive is especially interesting because it is the kind of information generated through actual business activity that is actually business-oriented rather than written for public consumption. Corporate conversations and operational records can be used to show how people communicate, solve problems and make decisions in real-life situations.
According to reports, the airline's systems contain records of more than seven billion flight transactions and extensive passenger and operational information. Such data is especially sensitive, and privacy protection and details of any transaction could be very important, so privacy protection and details of any transaction are key to this data, they say.
Privacy Could Become the Biggest Question
But there are serious privacy issues in the deal, the airlines say. Airline databases are heavy with sensitive information about passengers, employees and customers.
Some reports have indicated that personal information will be removed from the datasets before the information is provided for the reported transaction. Passenger-related data is said to include records of tens of millions of people, including loyalty-program information.
But removing personally identifiable information is only one part of the larger privacy issue. Large datasets can occasionally contain indirect clues that could be used to reconstruct identities or reveal sensitive patterns. The handling, anonymization and permitted use of the information would therefore be crucial.
The alleged Google-Spirit transaction is therefore about much more than a $10 million payment. It is indicative of an evolving digital economy in which billions of years of business data can be transformed into value as well.
The New Data Economy
From employees who walk around with cameras to take pictures of real-world environments for robotics research to companies that search through corporate archives for AI development papers, the methods used to acquire training data are growing more and more diverse.
The Spirit Airlines case offers another glimpse into that emerging market. A bankrupt company’s digital history– once viewed as an administrative archive– could be a useful resource in the race to develop better AI systems.
The bigger question here is whether companies and regulators are ready for a world in which corporate emails, software code, workplace conversations and historical databases can become valuable AI assets.
If the $10 million Spirit Airlines deal goes through, it could be an important example of that transformation. Planes may be one of the most visible assets of an airline but with AI, its most valuable possessions might be the digital records kept behind the scenes.
