Oracle Just Named AI in a Layoff Filing. White Collar Workers Should Pay Attention.
Sign Up For Daily Job Market News
The language in Oracle’s annual report is what makes this story different from the standard layoff announcement. The company wrote that “the adoption and deployment of AI technologies across our operations have resulted, and may continue to result, in reductions to our workforce.” For two years, companies cut staff and cited macro headwinds. Oracle put the actual reason in a regulatory filing.
The money went somewhere specific. Oracle expects to spend a net $70 billion this fiscal year on AI data centers, up from $55.7 billion the year before. Much of that buildout ties to a major arrangement with OpenAI, which is expected to purchase roughly $300 billion in computing capacity from Oracle over about five years. The company acknowledged that this financing will likely pressure margins.
Oracle also disclosed it may “fail to recoup” these investments if competitors’ AI products win the market or if costs run higher than expected. A 13% workforce cut to fund a bet the company itself flags as potentially unrecoverable. Workers in tech are navigating exactly that right now.
The CEO confidence data I covered earlier this month showed more CEOs plan to reduce headcount than grow it. Oracle is the clearest public example yet of what that looks like in practice. Meta and Amazon have also cut thousands of jobs in the same period, for the same stated reasons.
More Than Half of CEOs Are Eyeing White Collar Roles for AI Replacement
On June 11, the American Enterprise Institute and the Urban Institute jointly launched the Commission on AI and the Future of the American Workforce, co-chaired by former Commerce Secretary Gina Raimondo and former House Speaker Paul Ryan, with Google providing seed funding. A right-leaning and left-leaning think tank agreeing to work together on the same problem is itself a signal.
What matters is what they said out loud.
Raimondo framed it directly: “We have a tech strategy now. We lead the world with chips and models and innovation and companies. What I don’t think we have, in fact, I know we don’t have, is a people strategy.”
She put numbers to the risk. To be clear about how she framed them: these are warnings about what could happen, not forecasts she’s endorsing. Raimondo cited possible 25% unemployment among workers ages 25 to 35, double-digit unemployment for mid-career white collar workers, and “some estimates” putting 50 million jobs in the AI-vulnerable category. For scale, the China shock displaced roughly 3 million jobs, and the country is still working through the damage.
The detail employers should sit with: more than half the CEOs Raimondo speaks with are actively looking at their mid-level clerical, administrative, financial, and analytical roles and can see AI doing that work. These positions are the connective tissue of most organizations.
Ryan named the competitive trap those same CEOs feel. They’d rather augment their workers than cut them, but when a competitor might use AI to slash labor costs first, the pressure builds. “They’re worried that this could go to a game where the first who cuts labor costs wins,” he said, “but they don’t want to be that first.”
I’ve watched enough Washington commissions get announced and quietly archived. What’s different here is the specificity. Raimondo drew on her own family: her father losing his manufacturing job at 56 after 28 years when China entered the supply chain, and how he needed “a bridge to another chapter of work.” The “jobs we can’t imagine yet” optimism came up, as it always does, and it’s the weakest part of this conversation. A 45-year-old analyst whose role is disappearing this year can’t afford to wait for a future that hasn’t materialized. The bridge in between is where real people get hurt.
I’ve been tracking how the AI debate among the companies building this technology remains unsettled. Raimondo and Ryan are adding a different kind of weight to it: bipartisan policy credibility, and direct access to the CEOs making the actual headcount decisions.
Staffing your team doesn’t have to be hard.
Reach out and see how we can help.
Workers Are Faking Receipts With AI. For Some, It’s About Revenge.
An Atomik Research survey of 2,000 workers (1,000 U.S., 1,000 UK), commissioned by Emburse and fielded May 5 to 8, 2026, found that 40% of U.S. employees have used AI to generate a fake expense receipt: 19% completely fabricated a purchase, 15% inflated a real one, and 6% recreated a lost receipt for a genuine expense. Of those who faked receipts, 40% used their employer’s own AI tools to do it. Nearly 1 in 10 respondents overall built their own AI tools to generate fakes.
Financial stress is part of the picture. Among the 27% of U.S. workers who’ve passed off personal purchases as business expenses (up 3 points from 2024), 51% incurred overdraft fees or credit card interest waiting on slow reimbursements, up 11 points from last year. Slow reimbursement creates the conditions for this. Fix how quickly you pay people back, and you remove a significant chunk of the financial excuse.
The fraud number grabs headlines. The motivation underneath it matters more.
18% of respondents said they’re using company AI tools because they expect to be replaced and are taking advantage while they can. 22% are using employer-funded AI to apply for other jobs. When nearly 1 in 5 workers is operating in fear-or-revenge mode, the expense policy isn’t the root problem. The root problem is a workforce that’s already checked out.
I covered a related version of this distrust in last week’s Gallup data on AI and layoff sentiment. The direction is consistent.
