The May-2026 AI-skills-swap cohort gained its first auto-industry member on Monday, May 11.
General Motors confirmed it is eliminating approximately 600 salaried IT positions globally — more than 10% of its internal IT headcount — and concurrently opening requisitions for a narrower, AI-focused replacement skill set. The cuts are concentrated in Austin, Texas (the IT hub built up during the 2023 software-org expansion) and Warren, Michigan (GM’s historic engineering campus). Per Detroit News, CNBC, and Bloomberg, the global reductions began Monday and will affect 500–600 employees.
GM’s statement to Business Today was the cleanest dry-press-release of the week:
「GM is transforming its information technology organisation to better position the company for the future. As part of that work, we have made the difficult decision to eliminate certain roles globally.」
The replacement profile, per the same reporting, is the cleanest articulation of the Fidelity playbook yet to come out of the May cohort: AI-native software development, data engineering and analytics, cloud-based engineering, AI agent and model development, prompt engineering, and “new AI workflows.” Six job families on the hiring side. One job family — generalist enterprise IT — on the cutting side.
Why this looks like Fidelity in a different industry
The structural shape of GM’s announcement maps almost exactly onto the Fidelity May 7 / May 11 framing covered on this site: a named cut at a named scale, paired with an open hiring requisition pointed at the skills the cut wasn’t.
Two differences are worth noting.
First, GM is the only major US automaker in the cohort. Until May 11, the AI-skills-swap pattern had run almost entirely inside the financial-services and pure-software cohorts — Fidelity, Bill.com, Freshworks, PayPal, Coinbase, Cloudflare. GM is the first Detroit-style legacy industrial company to publicly publish a one-for-one IT-to-AI talent swap. Detroit’s IT departments were originally built to run ERP, CAD pipelines, dealer-network systems, and manufacturing execution. Those workloads aren’t disappearing. What’s changing is the company’s stated belief about which generation of engineer it wants running them.
Second, this is the second large IT-side cut at GM in 21 months. Per Business Today, GM cut roughly 1,000 software staff in August 2024 as it concentrated resources on “high-priority initiatives.” That cohort, too, was framed as a reorganisation rather than a downsizing. The 2024 cuts plus the 2026 cuts together amount to ~1,600 GM software/IT positions eliminated in the same window the company is publicly signaling expansion in autonomous-vehicle software, Cruise-restart planning, and onboard AI agents.
The Sterling Anderson context
The May 11 cuts land in the middle of a year-long software-org consolidation under Sterling Anderson, GM’s chief product officer since May 2025 (formerly head of Tesla Autopilot). Anderson has been merging GM’s previously separate technology units — vehicle software, infotainment, IT, autonomous-vehicle research — into a single chain of command. Several senior software executives have exited during the consolidation: Baris Cetinok, Dave Richardson, and Barak Turovsky, GM’s former Chief AI Officer.
The replacement hires lean younger and AI-native: Behrad Toghi joined recently as AI lead (ex-Apple); Rashed Haq joined as VP of autonomous vehicles. The shape is consistent: the layer of senior software leadership built up during the 2018–2024 vehicle-software push is being thinned at the top and at the individual-contributor level simultaneously, and replaced with a thinner, AI-skill-stack-native organization that reports up through a former Tesla Autopilot lead.
How this reads against the Gartner thesis
The Gartner May 5 finding, restated: in the cohort of $1B+ companies that piloted AI, 80% cut staff, but the cut rate had no correlation with deployment ROI. Gartner’s diagnostic was that real returns live in 「people amplification」 — using AI to make existing employees more productive — rather than in headcount substitution.
Fidelity’s announcement, structurally, is a people-amplification shape: cut 1,000, hire 5,300, net up. GM’s announcement, structurally, is closer to a true substitution shape with a wrinkle. GM has not published a net-hiring number against the 600 cuts. The reporting confirms IT hiring continues but for a narrower set of skills; it does not confirm that the new requisitions exceed the cuts. Until GM publishes a net figure — and the Cruise restart hiring plan is one possible offset — the company sits, on Gartner’s framework, in the bucket of cuts not paired with a stated amplification investment.
What to watch
- Whether GM publishes a net IT-headcount target for end-of-2026. Fidelity put a number on it (net up). The cohort’s no-ROI bucket is the one without a paired hiring number.
- Whether other Detroit automakers follow. Ford CEO Jim Farley’s white-collar warning from last year already telegraphed a similar move. Stellantis has not telegraphed it.
- Whether the 「prompt engineer」 requisition actually shows up on GM Austin’s careers page in the next 30 days, or whether the framing was a public-relations posture rather than an org chart.
- Whether the 600-cut number drifts up in subsequent reporting. The 2024 cohort started at “several hundred” and ended at ~1,000.
The cleanest read, on the morning after, is that GM has formally joined the May-2026 cohort that is using AI as the operating rationale for restructuring — but, unlike Fidelity, hasn’t yet committed to publishing the amplification side of the trade. The cut is announced. The hire is described in the abstract. The math is, for now, held back from the press release.