Selling AI as Job Cuts Is Making Enterprise Adoption Harder

This week brought two numbers that enterprise leaders should read side by side. A Pew Research Center survey found that 52% of American adults are now more concerned than excited about AI, and that 71% expect it to mean fewer jobs over the next two decades. Meanwhile, Ramp's spending data showed that nearly 56% of the businesses on its platform now pay for AI tools. Adoption is rising while trust among the people who will use these tools is falling. Our view is that the way many companies talk about AI, as a reason for job cuts, is a large part of that gap, and that it is starting to work against the adoption they are paying for.

What did this week's data show about AI and jobs?

Pew's findings, as summarized by The Register, show the share of adults more excited than concerned has halved since 2021, to 9%. Pew identified fear of job loss as the main driver. For the first time, adults under 30 are part of the concerned majority, at 55%, and 73% of them expect fewer jobs. These are the employees companies are counting on to become fluent AI users.

The backdrop is a year of layoffs framed around AI. Not every cut is about AI, but many announcements have named it. Through late May, companies had announced nearly 50,000 AI-linked job cuts this year, according to Challenger, Gray & Christmas data cited by CBS News.

Spending, meanwhile, keeps climbing. The share of Ramp's business customers paying for AI topped 50% in March and reached nearly 56% by July, TechCrunch reported, though Ramp's data covers more than 70,000 mostly smaller US businesses and not the whole market.

Why does framing AI as job cuts slow adoption?

Paying for AI is not the same as getting value from it. Value depends on employees using the tools well, redesigning their own work around them and telling leaders honestly where AI helps and where it fails. Each of those requires trust. An employee who believes AI exists to eliminate their role has every reason to use it minimally, keep their best workarounds to themselves and stay quiet when a pilot underperforms.

Practitioners at AI Connect events have made this point repeatedly. At COLLIDE 2025, Katie McDonald of UnitedHealth Group's Chief Data Office described an internal town hall where a comment about AI-generated code alarmed engineers, and said her organization measures AI differently: "We're not measuring the benefit of AI in headcount." The panel, on building trust and upskilling the workforce, argued that the people side decides whether AI programs succeed.

The evidence for large, AI-driven productivity gains is also thinner than the layoff announcements imply. A National Bureau of Economic Research working paper released this spring, based on a survey of chief financial officers, projected that AI-related job cuts in 2026 could run several times last year's pace even though the productivity gains have been slow to appear, Interview Query reported. Cutting first and measuring later risks removing the experienced people who know where the process breaks.

Isn't some AI-driven restructuring inevitable?

Yes, and it would be dishonest to pretend otherwise. Some tasks are being automated, and some roles will shrink. Companies face real cost pressure, and leaders who promise that no job will ever change will not be believed. Telling employees the truth about which work is changing is better than false reassurance.

The question is sequence and emphasis. Columbia's Daniel Keum told CBS News that the main effect so far tends to come through slower hiring, especially of junior workers, rather than mass replacement. That suggests most companies have more room than their announcements suggest to redeploy people as work changes, instead of leading with cuts and hoping adoption follows.

What should enterprise leaders do instead?

We think the companies that get the most from AI over the next two years will follow a different playbook:

• Measure AI by outcomes, not headcount. Track cycle times, error rates, revenue and customer results. If savings come, they will show up there, and the metric will not teach employees to fear the tools.
• Train in context. "Meaningful upskilling really comes from when it's contextual and not generic," Arpit Gularti, Associate Director of Analytics at The Clorox Company, said at COLLIDE 2025. Role-specific training shows employees what AI does for their own work.
• Ask people what they already use. Airia's Dave Horton told a AI Connect webinar on replacing shadow AI with sanctioned tools that employees will say how they use AI if they do not feel threatened. "I also think it's a signal of where employees see value in AI," he said.
• Commit to redeployment where possible. Be specific about how people whose tasks are automated will move into new work, and report on it.

AI adoption is a people problem as much as a technology one, a theme in our coverage of how AI can make leadership more human. This week's Pew numbers are a warning that the workforce is listening to how leaders talk about AI. Companies that want their AI spending to pay off should talk about it in a way their employees can believe in.

Frequently asked questions

What did Pew's August 2026 AI survey find?

52% of US adults are more concerned than excited about AI, while only 9% are more excited, and 71% expect AI to mean fewer jobs over the next 20 years.

Is AI causing layoffs?

Some 2026 job cuts have been attributed to AI, but researchers say the main effect so far tends to be slower hiring, especially of junior workers.

How can companies build employee trust in AI?

Measure AI by business outcomes rather than headcount, offer role-specific training and be specific about redeploying people whose tasks change.

Opinion pieces reflect the views of The AI Enterprise Magazine Team.

Filed under: People, Culture & Leadership, AI Strategy & ROI