“They ruthlessly cut costs, R&D, and employee benefits and then replace existing employees with overseas contractors. Innovation and growth take a back seat to sheer profitability.”
This is the operating manual that explains why IgniteTech’s much-publicized AI purge feels more like a familiar private-equity play. CEO Eric Vaughan’s account to Fortune, that IgniteTech “replaced nearly 80%” of its staff after an internal AI mandate and that changing minds “was harder than adding skills”, reads like the CEO’s justification for a radical reset. But Fortune’s language is vague: the story says employees were “replaced,” and Vaughan declined to disclose a specific number, which leaves open whether people were fired, converted to contractors, or outsourced.
IgniteTech is owned by ESW. For anyone who’s watched the ESW orbit, that vagueness is not accidental. ESW’s playbook, summarized in a long explanatory dossier that has circulated inside the industry, is blunt: buy distressed software, strip costs, move work to an hourly contractor model through a unit like Crossover (which has been described in Forbes as a “global software sweatshop”), and squeeze recurring revenue out of an existing customer base rather than invest in new products. The manual even describes moving headquarters, replacing local leadership, and instituting productivity spyware on remote contractors. The result, critics say, is a portfolio of patched-together products and a replaceable workforce.
Talk of AI is a convenient excuse for a strategy that predates the current boom. The Fortune piece recounts “AI Mondays,” compulsory sessions where staff had to work only on AI projects, and Vaughan’s description of the transformation as “existential.” But on Reddit and in employee forums the takeaway is less lofty: “Where software goes to die,” one commenter wrote, riffing on the company’s mantra. Others noted the company’s small customer base and questioned whether mass layoffs were simply a way to paper over a lack of demand.
A Glassdoor review points to the level of dysfunction at IgniteTech, “This is, without exaggeration, one of the most delusional companies operating today… Employees are forced to install invasive spyware on their machines… The company culture is one of paranoia and control.” That kind of firsthand color aligns with the ESW playbook’s emphasis on surveillance and hourly contractors, and it undercuts the narrative that this was purely a story about workers being too slow to learn.
Yes, AI is reshaping how firms allocate labor, but concrete examples show it’s often a pretext for margin moves rather than a pure technology story. Klarna publicly said an AI assistant was “doing the equivalent work of 700 full-time agents” after it pared hundreds of support roles; a move the company later walked back as customers complained and parts of the automation under-performed. Scale AI cut roughly 14% of its full-time staff and ended contracts with about 500 contractors this summer. These reductions came weeks after a major strategic investment and amid scrutiny of the company’s role in the GenAI supply chain. Duolingo’s CEO has been explicit about the limits of the “AI excuse”: Luis von Ahn told Fortune the company “has never laid off any full-time employees” and that contractor numbers fluctuate as automation changes the workload, not as a corporate crusade to reduce headcount.
Those are not isolated anecdotes. Outplacement firm Challenger, Gray & Christmas found more than 10,000 U.S. job cuts in 2025 where employers cited generative AI as a factor, while the tech sector’s total tracked cuts this year has already surged past 800,000: the highest since 2020. At the same time, research shows many of the layoffs tied to “AI” are concentrated in low-margin or outsourced roles rather than in core engineering teams. That intersects with an ownership model that prizes rapid margin improvement.
That context matters for how we judge IgniteTech’s claimed payoffs. The company points to patent filings and near-75% EBITDA margins as evidence the purge worked. But customers and former employees experience the other side: product fragmentation, opaque fee structures, and surveillance-heavy management. The Glassdoor review called the place “a masterclass in how not to run a company,” citing invasive tracking software and overbearing leadership.
If observers want to separate genuine AI-driven productivity gains from fiscal engineering dressed up as modernization, they should focus on three questions: who the cuts hit, which roles were automated rather than redesigned, and whether customers actually received better products. The public record on IgniteTech paints a different story to the narrative it is trying to sell.
