There is a particular kind of irony reserved for the company that sells you the future and then uses it on its own people first. ServiceNow spent the last two years telling every CIO on earth that its AI agents would let them do more with fewer humans. In late June it quietly proved the pitch by laying off several hundred of its own employees and explaining, with a straight face, that its own platform had made those jobs unnecessary.
”Real AI efficiencies,” and who paid for them
The cuts run into the triple digits — somewhere between a hundred and a thousand roles, per a source who spoke to Salesforce Ben. What makes this one notable is not the size. It is the framing. ServiceNow did not reach for the usual “restructuring” or “right-sizing” language. It said, more or less directly, that the layoffs are proof its product works: “Our platform is generating real AI efficiencies inside our own business.”
That is the marketing department writing the layoff memo. The same sentence is both a pink slip and a sales deck. Every customer evaluating whether ServiceNow’s agents can replace headcount now has a case study, and the case study is the vendor’s own staff. The people let go reportedly came from solution consulting, sales, product marketing, and learning and development — the functions, in other words, that explain and sell the platform, not the ones that build it.
The pledge that aged badly
The detail that gives this teeth: in 2023, while the rest of tech was running layoff after layoff, ServiceNow CEO Bill McDermott publicly pledged no job cuts. For a couple of years that held, and it was a genuine differentiator — the company that grew through the downturn without firing people. That promise is now retired. This is, per NowBen and HR Katha, the first layoff effort of the McDermott era.
The company’s stated logic is that it is “actively investing in and hiring for the AI-focused skills this era demands, while managing headcount with discipline to end the year where we started.” Read that carefully. “End the year where we started” means flat headcount — hire on one side, cut on the other. So this is not a company shrinking. It is a company swapping one kind of worker for another and calling the churn an efficiency. The net number stays put; the names on the badges change.
What this signals for everyone downstream
ServiceNow is not a struggling business reaching for a cost lever. It is a profitable, expanding one openly targeting $30 billion in revenue within four years. When a company in that position cuts staff and attributes it to its own AI rather than to any financial pressure, the message to the broader market is the uncomfortable one: this is not belt-tightening, it is the new operating model. The cuts are voluntary in the sense that the revenue did not require them. The AI did.
For anyone working adjacent to enterprise software — the consultants, the sales engineers, the people whose job is to translate a platform to a customer — the ServiceNow move is worth filing away. The first roles to go were not the deepest technical ones. They were the explainer roles, the human layer between a complex product and the buyer. That is precisely the layer a capable AI agent is built to thin out, and ServiceNow just demonstrated it on the friendliest test subjects available: itself.
The honest takeaway is not that AI ate these jobs in some clean, mechanical way. It is that “our AI works” has become the most flattering possible reason to cut, and a profitable company chose it on purpose. When the vendor uses the product on its own staff and then puts the result in a press statement, the demo is the layoff.
