99% of CEOs Plan AI Layoffs in the Next Two Years: Gartner Says 80% Won't Get the ROI

Your CEO has already decided to cut headcount tied to artificial intelligence (AI). The data says that plan is probably going to fail.

Two major research releases in late May 2026 landed within weeks of each other and, when read together, tell a story that every Chief Human Resources Officer (CHRO) needs to bring into the next executive meeting. The headline from Mercer's Global Talent Trends 2026 report is striking: 99% of CEOs expect AI and automation to shrink their workforce within two years. The counter-evidence from Gartner is just as stark: among companies that have already made those cuts tied to autonomous business deployments, roughly 80% are not seeing a return on investment (ROI) from those reductions.

That's the gap the CHRO has to close before the memo goes out.

The 99/32 Gap: Why Most CEOs Aren't Ready to Cut

The Mercer survey covers 825 C-suite executives and 1,650 HR leaders globally. The near-universal expectation of AI-driven headcount reduction is striking, but the number that matters more sits just below it: only 32% of those same C-suite leaders believe their organizations can effectively combine human and machine capabilities.

What the Data Says

• 99% of CEOs expect AI-driven headcount reductions in the next 2 years; only 32% feel their org can integrate humans and machines effectively (Mercer Global Talent Trends 2026, n=825 C-suite)
• Among organizations deploying autonomous business capabilities, roughly 80% report workforce reductions -- but those cuts don't translate to ROI (Gartner, May 2026)
• Employee concern about AI job loss rose from 28% in 2024 to 40% in 2026; only 44% report thriving, down from 66% in 2024 (Mercer 2026)

That 99-versus-32 spread is what the CHRO needs to print out and bring to the boardroom. It means the majority of CEOs plan to execute a headcount reduction strategy in an environment where they don't have the human-machine integration capabilities to justify it. They're planning to cut before the new operating model exists.

Mercer's data also shows that C-suite preparedness has dropped sharply. In 2024, 65% of executives said they felt well-prepared for the AI era. That figure is now 51%. At the same time, only 44% of employees report that they're thriving at work, compared to 66% just two years ago. Employee anxiety about job loss has jumped from 28% to 40% in that same window.

The picture isn't one of an organization confidently automating its way to efficiency. It's an organization where leadership confidence is slipping, employee morale is eroding, and the CEO's plan is still to cut.

Gartner's Finding That Should Stop the Layoff Memo

In early May 2026, Gartner published research on autonomous business deployments and arrived at a finding that should be uncomfortable reading for any executive planning workforce reductions as the primary AI ROI mechanism. Among companies that have piloted or deployed autonomous business capabilities, about 80% report they did reduce headcount. But those reductions didn't produce measurable returns.

The companies that did see ROI weren't the ones cutting aggressively. They were the ones investing in the skills, roles, and operating models that let people guide and scale autonomous systems. The cut-then-figure-it-out sequence doesn't work. The invest-then-optimize sequence does.

Fortune's coverage of the Gartner findings frames this clearly: AI layoffs can free up budget room, but freeing up budget and generating returns are two different outcomes. The organizations that mistake one for the other are going to have a very difficult conversation with their board in 18 months.

This is the conversation the CHRO can prevent from happening -- if the data gets surfaced before the decision is made.

What the 20% Who Get ROI Are Doing Differently

The Mercer data offers a specific clue about what separates organizations that will see returns from those that won't. Among all the AI initiatives executives could prioritize, 63% of respondents identified work redesign for AI automation as the highest-ROI initiative available to them.

Not headcount reduction. Work redesign.

This is the "Layoff-Before-Readiness Gap" in practice. An organization that reduces its workforce without first redesigning the work those people were doing hasn't automated anything. It's just smaller. The tasks don't disappear; they get redistributed, dropped, or done badly. The AI system that was supposed to handle the work often can't, because the workflow was designed around human judgment that no longer exists in the organization.

The 20% of organizations that do see ROI from AI deployments tend to follow a different sequence. They map the work first, identify where AI can genuinely take over well-defined tasks, redesign the roles around what's left, and then right-size the team to match the new operating model. The headcount reduction (if it happens at all) comes at the end of that process, not the beginning.

Mercer calls this "integrating human and machine optimally." Gartner operationalizes it as investing in roles that guide autonomous systems. The framing is different, but the direction is the same: you don't cut your way to AI ROI. You redesign your way there.

The CHRO Playbook for the Next 90 Days

If your CEO is in the 99% -- and statistically, they probably are -- here's a 90-day action sequence for the CHRO to redirect the conversation before the wrong decision gets made:

1. Build the 99/32/80 brief. Bring the three numbers together in a single slide: 99% plan to cut, 32% are ready to integrate, 80% who cut don't see ROI. These aren't your opinions. They're the aggregate experience of companies that have already been through this. The brief reframes the conversation from "should we cut?" to "are we in the 20% or the 80%?"

2. Map work before mapping headcount. Commission or accelerate a work design audit. For each function under consideration for AI-driven reduction, document what tasks exist, which are genuinely automatable with current tools, and what judgment calls require human oversight. This audit becomes the evidence base for any workforce decision.

3. Quantify the readiness gap. Mercer's 51% C-suite preparedness figure should prompt a diagnostic question: where does your organization sit? Run a rapid skills assessment across the functions closest to AI deployment. The gap is often not about technical capability -- it's about the judgment and oversight skills that make autonomous systems actually work.

4. Reframe the ROI conversation with the CFO. The CHRO and CFO need to be aligned before the CEO proposes cuts. The real question isn't whether AI will reduce some costs, but whether reducing headcount now (before redesign) creates costs that offset the savings. Severance, rehiring, institutional knowledge loss, and productivity dips during transition are real numbers.

5. Pilot the redesign-first model in one function. Choose a function where AI deployment is already underway or imminent. Run the redesign-first sequence explicitly: map the work, redesign the roles, upskill the team, then evaluate whether any reduction is warranted. Document the outcome. That case study becomes the CHRO's proof point for the next board conversation -- and the template for every other function.

What This Means for the Board Conversation

The CHRO's value in 2026 isn't in executing AI layoffs. It's in stopping the organization from making a decision that the data already shows doesn't deliver. Board members are sophisticated enough to understand the difference between budget relief and ROI. Most of them have read the same Gartner research.

The CHRO who walks into that conversation with the data and a redesign-first framework is doing the job. The CHRO who simply facilitates the headcount reduction without raising these questions is leaving the organization exposed to exactly the outcome Gartner documented: smaller, but not better.

The board conversation the CHRO should be driving isn't "how many can we cut?" It's "are we building the operating model that makes the cuts we do make actually stick?" That's a harder question. But it's the right one. And for the middle management layer that will carry out any redesign, getting the sequence right determines whether those leaders become AI accelerators or organizational bottlenecks.

The 20% who see ROI have already figured this out. The CHRO's job is to make sure their organization is not in the 80%.

Frequently Asked Questions

Should we delay AI deployment because of the readiness gap?

No. Delay isn't the answer. The readiness gap Mercer identifies isn't an argument against deploying AI -- it's an argument for sequencing the work correctly. Deploy AI in functions where you've already mapped the work and redesigned the roles. Don't delay deployment; delay the headcount reduction until the redesign is done. The upskill-vs-hire ROI framework can help prioritize where to start.

How do we explain to the board why headcount cuts alone won't deliver AI ROI?

Use the Gartner data directly. Among organizations that have already deployed autonomous business capabilities and made workforce reductions, roughly 80% are not seeing returns from those reductions. That's not a theoretical risk -- it's the observed outcome for most companies that have already run this experiment. The board can choose to be in the 20%, but that requires a different approach than cutting first and asking questions later.

What does "human and AI integration" actually look like in a CHRO operating model?

It means redesigning roles around what humans do better than AI (judgment, relationship, context, escalation) and what AI does better than humans (pattern recognition, volume processing, consistency at scale). In practice, it looks like fewer transactional tasks in every role, more oversight and decision-making tasks, and new roles that didn't exist before -- AI trainers, workflow supervisors, output reviewers. The measuring AI ROI framework gives a starting model for how to track whether that integration is actually delivering value.