GM Cuts 600 IT Jobs and Promises an AI-Powered Future — But the Math Doesn't Add Up

On the morning of May 11, 2026, hundreds of General Motors IT employees received short phone calls from their managers and HR representatives. By the end of the day, between 500 and 600 salaried workers — more than 10% of GM's IT department — had been told their positions were eliminated [1][2]. The company's explanation: their skills no longer fit. GM said it needed to bring in workers with expertise in artificial intelligence, agent development, and model engineering [2].

The cuts hit hardest in two locations: Austin, Texas, and Warren, Michigan [1][3]. Affected roles spanned infrastructure, quality assurance, and traditional software engineering — the back-office functions that keep a 118-year-old automaker's digital systems running [4].

GM characterized the move not as a reduction but as a transformation. "As GM continues to build a more modern, efficient, and innovative technology organization, we are adjusting our team to align with the skills and capabilities we need for the future," a spokesperson told reporters [1]. The company pointed to 83 open IT positions on its careers site, many involving AI, motorsports, and autonomous vehicle technologies [2].

But 83 openings to replace 600 eliminated jobs is not a one-for-one swap. It is a net cut of more than 500 positions.

The Scope of What Was Cut

The affected workers were salaried employees, not hourly union workers covered by GM's UAW contract. Their roles included quality assurance engineers, IT infrastructure specialists, and software engineers working on internal systems [4][5]. According to compensation data from Levels.fyi, GM QA engineers earned between $96,000 and $132,000 annually, with a median of about $115,000. The overall median total compensation for GM IT staff was approximately $133,000 per year [6].

Source: Levels.fyi, Glassdoor, industry data

Data as of May 12, 2026CSV

The AI-focused roles GM says it wants to fill carry substantially higher market rates. AI engineers command average total compensation of $185,000 nationally, with senior machine learning engineers at top-tier companies earning $300,000 or more [7]. This creates an apparent paradox: if GM is replacing lower-paid workers with higher-paid specialists, the move is not primarily about cost savings. But if GM is only hiring 83 people to replace 600, the net labor cost reduction is significant — roughly $60 to $70 million annually at median compensation levels, before accounting for benefits and overhead.

GM has not disclosed the projected savings from this restructuring or broken out how many of the 83 openings are net-new versus backfills.

A Pattern, Not an Isolated Event

These IT layoffs are the latest in a series of workforce reductions at GM. In late 2025, GM permanently laid off more than 1,100 workers at its Factory Zero assembly plant in Detroit [8]. Around the same time, over 2,000 workers were placed on temporary or indefinite layoff at GM's Ultium Cells battery joint ventures in Lordstown, Ohio and Spring Hill, Tennessee [8]. In October 2025, GM cut hundreds of additional salaried workers as part of what it called a "profit push" [9].

Source: Company announcements, news reports

Data as of May 12, 2026CSV

GM is not alone. The automotive sector announced 45,820 job cuts through November 2024, a 59% increase from the prior year [10]. Ford eliminated 3,000 positions in early 2024, primarily in software development and engineering roles across Europe and India [11]. Stellantis cut 400 white-collar technology and engineering workers as part of its own EV transition cost-cutting [12]. The entire industry is shedding headcount as EV demand falls short of projections and traditional revenue streams face margin pressure.

The AI Rationale: Strategic Pivot or Corporate Cover?

The timing of GM's "skills swap" narrative deserves scrutiny. An Oxford Economics report published in January 2026 argued that AI-related layoff explanations are increasingly functioning as "corporate fiction" [13]. The research found that AI was cited as the reason for nearly 55,000 U.S. job cuts in the first 11 months of 2025 — but this represented only 4.5% of total reported job losses. By comparison, layoffs attributed to standard "market and economic conditions" were four times larger, at 245,000 [13].

Oxford Economics concluded that attributing staff reductions to AI "conveys a more positive message to investors" than admitting to cyclical business pressures or past over-hiring [13]. Wharton professor Peter Cappelli observed that corporate announcements often say "We expect that AI will cover this work" — expressing hope rather than confirmed capability — "because that's what they think investors want to hear" [13].

GM's own financial context reinforces this reading. In January 2026, the company recorded a $6 billion write-down related to its EV business after slower-than-expected adoption and higher-than-anticipated variable costs [14]. CEO Mary Barra has emphasized "capital discipline" and "prioritizing profitability and flexibility" during earnings calls [2]. The company is targeting approximately $2 billion in cost reductions to offset tariff exposure projected at $5 billion [14]. Against this backdrop, an IT workforce reduction framed as an AI upgrade serves dual purposes: reducing costs while signaling forward-looking investment to Wall Street.

What the Research Actually Shows About AI and Layoffs

A Gartner study published the same week as GM's layoffs — based on a survey of 350 global business executives at companies with at least $1 billion in annual revenue — found no correlation between AI-driven workforce reductions and improved return on investment [15]. Workforce reduction rates were "nearly equal for those reporting higher ROI and those with smaller returns or even worsened outcomes," the study found [15].

Helen Poitevin, a Distinguished VP Analyst at Gartner, stated: "Many CEOs turn to layoffs to demonstrate quick AI returns; however, this disposition is misplaced. Workforce reductions may create budget room, but they do not create return" [15]. The study found that companies with the highest gains were those using AI for "people amplification" — making existing workers more productive — rather than replacing them outright [16].

A separate NBER survey of 750 U.S. chief financial officers projected that AI-driven job cuts in 2026 would reach approximately 502,000, roughly nine times the approximately 55,000 reported for 2025 [17]. But critically, the survey noted that many of these projected cuts were "preemptive, driven by CFO expectations about AI's future capabilities rather than roles already replaced by working systems" [17]. Companies are restructuring around a future that has not fully arrived.

Goldman Sachs senior economist Ronnie Walker put it plainly: "We still do not find a meaningful relationship between productivity and AI adoption at the economy-wide level" [17].

The Reskilling Promise — and Its Track Record

GM said it is "committed to supporting [affected workers] through this transition" [1], but the company has not publicly detailed what retraining, severance, or placement assistance is being offered. This matters because corporate reskilling promises have a poor historical track record.

According to McKinsey, executives estimate about 40% of their workforce needs to reskill over the next three years. Yet only 6% of companies surveyed said they had begun upskilling workers "in a meaningful way" [18]. Just 26% of organizations have conducted a skills audit to assess AI's impact on roles, and only 30% are investing in reskilling and redeployment pathways for at-risk positions [19].

The track record of workers laid off during prior technology transitions being rehired by the same employer is thin. Public data on GM-specific rehire rates after previous restructurings is not available, and GM did not respond to questions about whether any of the 600 laid-off IT workers were offered the opportunity to apply for the 83 open AI-related positions with reskilling support.

The Geographic and Human Impact

The concentration of layoffs in Austin, Texas and Warren, Michigan affects two communities with very different labor market characteristics. Warren, a Detroit suburb, is deeply tied to the auto industry and has experienced repeated waves of GM restructuring over decades. Austin's tech economy offers more alternative employment options, but GM's IT presence there was itself a product of earlier expansion meant to tap into the city's technology talent pool [1][3].

Source: FRED / Bureau of Labor Statistics

Data as of Apr 1, 2026CSV

The national unemployment rate stood at 4.3% as of April 2026, up from 3.5% in mid-2023 [20]. While this remains historically moderate, the upward trend — and the concentration of layoffs in specific sectors — means displaced IT workers face a tighter market than the headline number suggests. Yale researchers have found that AI's employment impact falls disproportionately on entry-level and mid-career workers, who find it harder to secure new positions as companies freeze hiring for roles they expect AI to eventually handle [21].

Legal Considerations

The federal Worker Adjustment and Retraining Notification (WARN) Act requires employers with 100 or more full-time employees to provide 60 days' written notice before mass layoffs affecting 500 or more workers at a single site, or 50 or more workers at a site if they constitute at least one-third of the workforce [22]. Given that GM's cuts reportedly affected 500 to 600 employees across at least two locations, the WARN Act's applicability depends on the distribution of layoffs per site.

No public WARN Act filings related to these specific cuts had appeared in Texas or Michigan state databases as of publication. No lawsuits or formal complaints alleging the AI-skills rationale was pretextual have been reported, though affected workers may still be within the filing window for such claims.

GM's salaried IT workers are not covered by the UAW collective bargaining agreement, which limits the union's ability to intervene on their behalf [4].

The Case For — and Against — Early AI Adoption

The strongest argument in GM's favor comes from economic research suggesting that companies that delay AI adoption may shed more jobs over a longer period than those that restructure early. A Carnegie Endowment paper published in April 2026 identified three schools of thought on AI and labor: substitution (AI replaces workers), augmentation (AI makes workers more productive), and reshuffling (AI eliminates some roles while creating others) [23]. The reshuffling view — which most mainstream labor economists hold — suggests that companies that proactively restructure around AI can position their remaining workforce for the new roles that emerge.

Morgan Stanley research has characterized AI's labor market impact as likely to "echo past tech transitions," with modest near-term displacement followed by new job creation in adjacent areas [24]. The a16z venture capital firm has called predictions of an AI job apocalypse "unhelpful marketing, bad economics and worse history" [25].

But the counterargument is equally data-driven. The Gartner study suggests that the specific mechanism GM is using — layoffs followed by selective rehiring — is less effective than retraining and redeploying existing workers [15]. And the Oxford Economics data indicates that the AI narrative is being used selectively by companies that are primarily responding to financial pressure, not technological capability [13].

What GM Is Actually Building

The restructuring has a clear organizational architect. Sterling Anderson, co-founder of autonomous trucking startup Aurora, was hired as GM's chief product officer in May 2025. Since his arrival, three top executives have left GM's software team as Anderson has worked to consolidate the company's disparate technology operations into a single organization [2]. GM also hired Behrad Toghi, formerly of Apple, as AI lead in October 2025 [2].

The capabilities GM says it is hiring for — agent development, model engineering, AI-native workflows — represent a real and measurable shift in the kind of work the company's technology division is doing [2]. Whether this shift required eliminating 600 positions or could have been achieved through gradual attrition and retraining is the question GM has not answered.

Mary Barra has pointed to concrete AI applications: GM reported a record 12 million OnStar subscribers in 2025, including more than 120,000 Super Cruise advanced driver-assist users, with the service expected to generate $400 million in high-margin revenue in 2026 [14]. The company's roadmap targets "eyes-free" advanced driver-assist technology by 2028 [14]. These are real products with real revenue, not speculative applications.

The Bottom Line

GM's IT layoffs are simultaneously a genuine organizational restructuring, a cost-cutting measure during a period of financial pressure, and a signal to investors that the company is embracing AI. These explanations are not mutually exclusive — which is precisely what makes the "AI skills swap" framing misleading. The company is replacing 600 workers with 83 openings, booking tens of millions in labor cost savings, and presenting it as a capability upgrade.

The broader context is equally important. A survey of CFOs projects 502,000 AI-related job cuts across U.S. companies in 2026 [17]. Yet the best available research shows no macroeconomic evidence that AI adoption is improving productivity at scale [17], and the companies that cut workers in the name of AI are not outperforming those that retrain them [15].

For the 600 workers who lost their jobs on a Monday morning in May, the distinction between strategic transformation and cost reduction is academic. Their severance terms, their access to retraining, and their prospects in a labor market increasingly shaped by AI anxiety will determine whether GM's promises of a more "modern, efficient, and innovative" future apply to anyone beyond the company's balance sheet.