Half of Americans fear AI could put someone in their household out of work, according to a Reuters/Ipsos poll. If you are job-hunting right now, or weighing an offer from a company that just made an AI-driven restructuring announcement, that fear is not paranoia. New research from the University of Pittsburgh finds that the productivity story executives tell about AI-linked layoffs routinely diverges from the results their own companies are producing, and the workers left behind are noticing.

Mark Ma, Professor of Business Administration at the University of Pittsburgh, led a study that analyzed millions of job-satisfaction reviews, thousands of corporate financial reports, and hundreds of AI-linked layoff announcements by U.S. public companies over the past five years. The pattern that emerged is worth understanding before you sign anything.

The productivity story executives tell does not match their own results

An Atlanta Federal Reserve survey found roughly 90% of executives believe AI has not yet boosted productivity at their companies. Ninety percent. That is the same group making confident AI statements on quarterly earnings calls.

Mark Ma and his team went further, pulling around 10,000 earnings-call transcripts and measuring how management discussed AI. The tone was consistently optimistic. But that optimism had no significant relationship to actual productivity outcomes at those same companies. In plain terms: the more confidently a CEO talked about AI on a call, the less that confidence predicted whether the business would actually get more productive.

For you as a candidate, this is useful context. When a recruiter describes a company as AI-first or says it is restructuring for the AI era, that phrase may be brand positioning rather than a business plan with measurable targets attached. You are fully entitled to ask what it means in practice for the role you are considering.

What AI-justified layoffs really look like from the inside

The University of Pittsburgh research surfaced one detail worth sitting with: some companies in the study made layoffs before investing in AI, using the payroll savings to free up capital for future AI spending. The AI justification appeared in the press release first; the actual technology investment came later. That is a specific pattern, and it means an AI restructuring announcement can sometimes be a cost-cutting decision with a cleaner narrative attached to it.

Meanwhile, the employees who kept their jobs were not quietly settling in. Workers leaving AI-related comments on Glassdoor were markedly more negative than the overall tone of reviews on the same platform. The Bureau of Labor Statistics (BLS) Job Openings and Labor Turnover Survey (JOLTS) has consistently shown that voluntary separations rise in sectors where workers sense instability, which aligns with what the Pittsburgh study found internally. The complaints from those AI-tagged reviews clustered around a few themes:

• Job-security concerns, by far the most commonly cited.
• No appropriate training on the AI tools being introduced.
• Few chances to upgrade skills before being expected to use the new systems.
• Poor corporate AI leadership, meaning the strategy felt improvised rather than planned.

If you are interviewing at a company that recently announced AI-linked cuts, those Glassdoor signals are an early-warning system. They tell you what the people who stayed think of the plan, not what the press release says.

The market shrugs, even when the company brags

If AI-driven layoffs were genuinely creating lasting value, you would expect investors to reward them. According to the Pittsburgh research, they mostly do not. Stock-market reactions to AI-linked layoff announcements averaged close to zero, and more than half of these events produced negative or near-zero returns.

There are exceptions. Block, the financial-tech platform, saw its share price jump when it announced it would trim staff on AI grounds. That kind of headline-grabbing result gets quoted repeatedly. But on average, the market treats a company cutting jobs because of AI as a non-event. That is a quiet but concrete verdict on how much of the AI-layoff story reflects real strategy versus corporate theater.

The Economic Policy Institute (EPI) has noted separately that productivity gains from automation have historically taken years to materialize at the firm level, and that early-mover announcements often signal intent rather than demonstrated results. The Pittsburgh findings fit that longer historical pattern.

What this means when you are comparing offers

You cannot outmaneuver a bad restructuring by sheer willpower, but you can read the signals before you sign. A few questions worth raising in a final-round interview, or reading between the lines of before you accept:

• Has the company announced AI-related layoffs in the past 18 months? If yes, ask what happened to the team you would be joining. Expanding this function after a broader cut signals something different than rebuilding this team after cuts.
• Is the AI investment concrete or aspirational? A company that names specific tools, teams, or workflows using AI is in a meaningfully different place than one repeating buzzwords on a loop. Vague answers are data too.
• What training is offered on the AI tools you would be expected to use? The Pittsburgh research flagged no appropriate training as one of the loudest complaints from workers already inside these companies. Ask early, not on day 90.
• How are performance expectations changing? If AI is supposed to make you more productive, output targets often shift alongside it. Understand what more productive looks like on your future manager's dashboard before you accept the role.
• What do current employees say? Reviews on independent platforms are noisier than official communications but often more honest. Read the AI-tagged reviews specifically, not just the headline score.

None of these questions make you sound defensive. They make you sound like a candidate who reads the market carefully. Companies that push back on reasonable due-diligence questions are giving you information too.

Reading a job ad in the age of AI restructuring

Some job ads now read like technology manifestos. That is usually about brand positioning, not the role you would actually do day to day. When you see phrases like AI-native, automation-first, or post-restructuring rebuild, ask what changed in the past year and who is left on the team. If the job spec was rewritten after a layoff, the workload description may not reflect what the previous person in the seat actually did.

Compare the ad to public reviews, headcount trends, and the company's own earnings-call language if it is a public company. When executive tone is uniformly optimistic but current employees flag training gaps and unclear leadership, the employee reviews are the more reliable signal. The Pittsburgh study found the two versions of reality diverge more often than not.

Your rights and options if you are already inside

If you are an employee at a company that has just announced AI-linked changes, a few practical moves apply right now under the current U.S. framework.

• Know your WARN Act protections. Under the federal Worker Adjustment and Retraining Notification (WARN) Act, employers with 100 or more employees are generally required to give 60 days advance written notice before a mass layoff or plant closing. If a restructuring crosses that threshold and notice is not given, you may be entitled to back pay and benefits. Some states, including California and New York, have their own WARN-equivalent statutes with broader coverage, so the federal floor is not always the ceiling.
• Get the plan in writing. Restructuring communicated only verbally is harder to hold anyone to. Ask for the timeline, the roles affected, and the training plan by email so you have a record.
• Document your current output. If AI tools are being introduced and productivity expectations shift, having a clear baseline of what you delivered before the tool arrived protects you in later performance conversations.
• Take the training the company offers, and log it. Even if the program is thin. It is evidence of engagement if roles get reviewed later and someone is comparing records.
• Check your severance agreement terms early. Severance is not federally mandated under the Fair Labor Standards Act (FLSA), but many employers offer it under company policy or negotiated agreements. Under the Employee Retirement Income Security Act (ERISA), severance plans run through a formal plan document may carry additional protections. Know what yours says before you need it.
• Keep your resume current. Not as panic, but as posture. The workers who navigate restructurings best are usually the ones who could leave on their own terms if they had to.

The quieter takeaway

The University of Pittsburgh research does not argue that AI will not change work. It clearly will. What it shows is that the specific move of announcing layoffs and attributing them to AI has, so far, not produced the productivity boost executives claim to expect, has not moved stock prices on average, and has damaged worker sentiment in measurable ways. That is a consistent gap between the pitch and the outcome.

For you, the practical version is straightforward. When a company's AI story sounds louder than its results, treat the story as marketing. When employees inside that company describe thin training and improvised leadership, treat those reviews as the more honest data. Mark Ma's research at Pittsburgh gives you the framing; verified employee reviews on platforms like wherewework give you the company-specific signal. Your next role is a decision, not a headline. Ask the questions that make that decision easier to defend to yourself in twelve months.