Mark Ma wants to know how the workforce really feels about artificial intelligence — so he’s tracking exactly that.

An associate professor of business administration at the University of Pittsburgh, Ma spent the past four years studying the motives behind return-to-office mandates after the pandemic. He became a go-to voice on the issue after publishing a study with Ph.D. student, Yuye Ding, that argued return to office mandates don’t improve employee or company performance.

Now, he, Ding and a team of researchers have created the AI Sentiment Tracker.

Using data from Glassdoor job postings, corporate earnings calls, layoff announcements and other market statistics, Ma and his team are weaving an AI story at the individual, company and regional levels. The study has analyzed more than 3,200 firms across “six key dimensions — AI talent share, hiring, retention, salary premiums, and employee and executive sentiment,” according to the tracker.

In an interview with the Pittsburgh Post-Gazette, Ma contemplated the future of the workforce as the push for AI adoption goes full throttle.

This interview has been edited for length and clarity.

Q: Is the concern over losing jobs real? Should we all be scared about losing our jobs to AI, or is it overhyped?

A: I think it is very real, in the next five to 10 years.

Very soon, similar to all the other technologies, the cost of adopting AI will go down. Now, it’s still quite high in many professions to use AI rather than human beings, but the cost will come down, and more and more companies can adopt AI.

If management’s beliefs do not change, or their strategy does not change, we will see more and more companies trying to lay off people and adopt AI.

At the same time, there is an argument that AI will create new jobs — but it’s really unclear what new jobs will be created, not be replaced, by AI.

If you ask Nvidia’s CEO, Jensen Huang — and he is one of the best CEOs who treats employees so well — but if you ask him what new jobs will be created, he will tell you to use your imagination.

He doesn’t really have a solid, concrete answer. He is the leader of the revolution, and he doesn’t even know the answer.

Q: Do you think AI will create new jobs?

A: I think the new jobs that will be created will be fractional to the jobs that we will lose.

A lot of people are saying that you should start looking for trade jobs, like plumbers, electricians. But if all these laid-off people get trained and become electricians and plumbers, the supply of plumbers and electricians will increase a lot. Will that make the plumbers’ market even more competitive?

We will have to make some public policy changes, but that will only happen after the worst case scenario. That’s just how things work.

Policy changes never happen before the worst case happens. Only after we have this huge problem, and all these people get so fed up, and they lose their job, can’t afford paying their mortgage — maybe that’s what will happen. If people can’t pay their mortgages, the financial institutions will urge those politicians to make changes.

I’m optimistic that we will reach a new equilibrium where AI benefits the society in the long term, but the path to that destination, I think, will be very, very bumpy.

Q: How long do you think it will take to get to that utopia?

A: I think the massive layoff will come within five to 10 years, and after that they will make changes.

Then all the benefits will start. I just hope that we will be prepared.

The most ideal case is you continuously track all these unemployment issues, and then you develop a long-term plan. Maybe after unemployment reaches 10%, you decrease the work week from 40 hours to 35. You gradually reduce your labor supply as labor demand decreases, so there is a new equilibrium. (That way) you could keep the same wage level, so everyone could be paid under the same rate, and we could keep full employment.

Q: What does your sentiment tracker do?

A: We try to capture the sentiment from multiple aspects, like individual level, company level, regional level, and economy level.

First, we look at the employees — who are really at the center of this revolution. They are asked to use AI at the same time they are being replaced by AI, so they are the group that are most affected by AI. We want to see what they think about AI.

In almost every case, (employees’) sentiment of the AI-related issue is much more pessimistic than the overall sentiments towards the firm. That means they are still satisfied with the firm, but they are really not happy with the AI adoption.

The most negative issue is job security concerns. They are concerned that the companies are asking them to use AI, and once the company efficiently adopts the AI, they will be replaced. Because they have been observing that the layoffs become a part of the company’s strategy.

They are kind of competing with each other about who will survive.

Q: What about the managers?

A: Managers are very optimistic, but the employees are not so optimistic.

At the firm level, there is a huge gap between manager and employee, and it seems the managers are still not fully understanding the situation. They are still very optimistic that AI will bring huge benefits to the firm, but that benefit is not materializing under many companies.

We also analyzed whether the management sentiment and employee sentiment actually explain firm productivity.

We found the employee sentiment significantly explains firm productivity, but management sentiment does not really correlate with the firm’s productivity. That basically means the negative sentiment by employees is hurting productivity.

Managers were expecting AI would improve productivity, but because of job security concerns, and also maybe the lack of appropriate upscaling opportunities, that benefit is not materializing in most companies.

Q: What advice would you give to students and the workforce?

A: I think you should spend a lot of time practicing your AI skills in today’s market.

Many students are protesting against AI, and I think they should; but public policy change often comes much later, after the worst case of unemployment happens.

On the one hand, you want to voice your concern. On the other hand, you want to practice using AI tools and become AI-skilled, so you maximize your competitiveness in the job market.

We cannot delay the AI adoption. ... In the future, you will have to use these tools.

©2026 PG Publishing Co. Visit at post-gazette.com. Distributed by Tribune Content Agency, LLC.