Newsrooms are supposed to scrutinize powerful companies. It gets complicated when one of those companies becomes a vendor.
More than 800 USA Today Co. employees are now calling on their employer to end its recently announced partnership with Palantir, according to a joint statement from 31 unionized newsrooms reported by Nieman Lab.
The dispute centers on USA Today Co.'s decision to use Palantir's software to analyze and monetize audience data across its publishing operations.
For journalists, however, the partnership raises questions that go well beyond the technology itself.
Why Journalists Are Objecting
USA Today Co., formerly known as Gannett, announced the Palantir partnership during its second-quarter earnings call on August 6.
The company said the technology would help it “strengthen how we collect, connect and activate audience data to drive more effective and faster monetization across our platform.”
That business case is straightforward. News publishers are under enormous pressure as search traffic declines and traditional advertising models weaken.
The newsroom response has been anything but straightforward.
In their joint statement, employees argued that the partnership could undermine trust in USA Today outlets while raising questions about reader data security and the company's relationship with a controversial technology provider.
The employees also point to a potential editorial conflict: Palantir is itself a company that journalists may need to cover.
“Not only does partnering with Palantir, a major player in the news we cover, create an inherent conflict of interest,” the statement said, “but it fails the journalists who have been assaulted and falsely arrested while covering immigration enforcement actions and protests.”
The Data Question
One of the biggest concerns is how Palantir's software will actually be used.
USA Today Co. says it will retain ownership of its audience data and that Palantir's technology will help the company make better use of it.
But employees say they received very little information about the arrangement.
Mike Davis, a reporter for the Asbury Park Press and vice-chair of the NewsGuild of New York, told Nieman Lab that employees learned about the partnership in an email roughly 20 minutes before the company's earnings call.
“Part of the problem is that we know very, very little about how Palantir’s software is going to be used,” - Davis
That lack of detail is particularly sensitive inside a newsroom.
Investigative journalists can hold confidential source information, unpublished reporting and sensitive documents on company devices. Employees therefore have a different relationship with data security than a typical corporate workforce.
The concern isn't necessarily that Palantir will gain access to journalists' confidential material. The concern is whether employees can be confident about exactly what information the technology can access and how those boundaries will be maintained.
Palantir's Reputation Matters
The controversy would probably look very different if USA Today Co. had chosen an ordinary enterprise analytics vendor.
Palantir's business has long been closely associated with defense, intelligence, immigration enforcement and large-scale data analysis.
That history makes the partnership particularly sensitive for journalists who regularly cover government surveillance, immigration enforcement and civil liberties.
Employees cited what they described as tensions between USA Today Co.'s stated commitment to First Amendment principles and Palantir's controversial government work.
For Palantir, meanwhile, partnerships such as this demonstrate how its technology is expanding beyond its traditional government and defense markets into mainstream commercial applications.
That creates a strategic benefit for the company: the more ordinary enterprise customers adopt Palantir, the less exclusively it is perceived as a defense and intelligence contractor.
The Company's Response
USA Today Co. has defended the arrangement.
The company told Nieman Lab that its editorial decisions remain independent and said its audience data is protected through existing privacy commitments, contractual safeguards and applicable laws.
“Our journalism remains fully independent,” Amy Garrard, the company's vice president of labor relations, said in a statement.
That distinction is important.
The dispute is not evidence that Palantir has editorial control over USA Today journalism. Rather, it is an argument from employees that the commercial relationship itself creates risks to trust, privacy and perceived independence.
Those are different questions, and the second may prove harder for media companies to resolve.
A Bigger Problem for AI Vendors
The USA Today dispute points to a broader issue emerging as AI becomes embedded throughout large organizations.
Companies increasingly want AI systems that can analyze customer behavior, automate workflows and extract value from huge datasets.
But choosing an AI vendor isn't purely a technical decision anymore.
Boards and executives also have to consider a vendor's reputation, government contracts, data practices and the potential reaction from employees, customers and regulators.
For media companies, the stakes are even higher because credibility is part of the product.
A publisher can argue that an AI system is secure and that editorial decisions remain independent. It still has to convince readers and journalists that those safeguards are meaningful.
What This Means for Miami
Miami's growing enterprise AI sector includes companies working across finance, healthcare, logistics, defense-adjacent technology and media. Many operate in industries where sensitive data and institutional trust matter as much as technical performance.
The USA Today dispute offers a useful warning for South Florida companies evaluating AI vendors: procurement decisions can create reputational consequences long after the technology has been deployed.
For Miami's fintech, healthcare and media companies, vendor due diligence should therefore extend beyond model performance and pricing.
Who else does the vendor work for? What data can it access? How is that data protected? And how will employees and customers perceive the relationship?
As AI becomes infrastructure rather than an optional experiment, those questions are becoming part of the technology decision itself.
