AI Fears are Hitting Consumers

Photo by christian koch on Unsplash

I was at a conference in Washington D.C. this week, and the morning keynote on the first day was given by an economics professor who specializes in AI. I am paraphrasing slightly because I wasn’t taking hard notes, but he said, essentially, “The economic value of cognitive knowledge in 5-10 years will be zero.” All of us very knowledgeable people in the room shifted in our seats uncomfortably.

The premise is this: AI technology is accelerating so rapidly that any cognitive task will be able to be completed faster and with more accuracy by bots than by humans in the very-near future. Any cognitive task. The human brain may be able to master a particular topic in, say, a year (and “master” is a loose term here), but AI can master almost any subject area in a matter of hours or perhaps a few days of complex calculations. OpenAI is said to be soon rolling out a $20,000/month AI agent that can work for your business, academic institution, or non-profit organization to the level of a PhD scientist.

I am in the early stages of drafting a two-part article series on 1) the most optimistic case for future AI impact on our country and world and 2) the bleakest case. So let me put a pin in those parts of the conversation for now, and we’ll come back to them soon.

What I am seeing anecdotally in my conversations with regular people, and what is beginning to hit the actual quantitative data on consumer confidence and American spending habits, is that people are starting to become fearful of the impact of AI on their jobs, and therefore on their own economic security. According to a recent report from Mercer, “Employee concern about job loss due to AI has surged from 28% in 2024 to 40% in 2026.” I predict that this level of concern will double again by 2027.

This comes even as there are relative signs of strength in the labor market. For example, wages were up about 3.8% in 2025. While that increase is nothing to particularly write home about, this level of wage growth did actually outpace inflation by about 1.0% last year. So on net, workers gained a bit relative to the cost of overall goods and services last year. This was not the case from 2021-2023, a period of time in which inflation generally outpaced wage growth; in those years, workers lost ground relative to costs (particularly in crucial areas like rent, housing, groceries, and utilities). So the fact that wages outpaced inflation in 2025 (they also did, modestly, in 2024), is a good thing.

The unemployment rate also remains at historically attractive levels, so people who want to work generally are able to find it (although maybe not at the wage level they want or need, to be fair). When you combine modestly rising wages with a low unemployment rate, it suggests a labor market that is relatively stable. But this is not what people are feeling.

Why? I think one main reason is that although wages did outpace inflation in 2025, people are still reeling from the inflation-fueled 2021-2023 period. Costs haven’t come down in any sort of meaningful way, and it takes a while for people’s psychological relationship with the cost of, say, a week’s worth of groceries to become re-anchored to the new income-price relationship.

But I think a big part of the story right now are fears about AI. People read the news, they see stories of layoffs starting at big tech companies, and they are starting to see AI implementation in their own workplaces. Consider the following:

• A recent Forbes report found that 75% of workers are concerned about AI causing job losses within the next 12 months. “These results suggest that, as AI becomes more integrated into various industries, the fear of job displacement presents a significant concern,” the report says.

• A Pew Research report found that people are more worried than hopeful about AI, with only about 6% of poll respondents thinking that AI will provide them with more opportunities in the workplace.

While the monthly University of Michigan consumer sentiment report does not ask respondents specifically about their thoughts on AI, there are some clues. The current readings from the report sit near all-time lows. Since the survey began in 1946, there have been only three times when the consumer sentiment reading was lower than it is today: 1) during the 1980 period of extremely high interest rates and a tough economic recession, 2) during the 2008 financial crisis, and 3) during the 2022 period of maximum recent inflation. Consumers are not feeling good right now, especially in the middle and lower parts of the wage/spending spectrum.

The January 2026 Michigan report notes that while consumer sentiment is up modestly from six months ago, it is down significantly (about 20%) from a year ago, which is a big decline in one year. The report says:

On net, modest increases in current personal finances and buying conditions for durables were offset by a small decline in long-run business conditions. While sentiment is currently the highest since August 2025, recent monthly increases have been small—well under the margin of error—and the overall level of sentiment remains very low from a historical perspective. Concerns about the erosion of personal finances from high prices and elevated risk of job loss continue to be widespread.

I’ve added the bold emphasis on the last line, which I believe is indicative of AI fears. As mentioned, there have been plenty of stories in the news over the past six months about AI and the automation and consolidation of jobs, and people in all walks of life are fearful right now that their jobs could be eventually eliminated. If the economist who I heard speak earlier this week is correct, many of us are right to be afraid of what the labor market will look like not just 5-10 years from now, but perhaps even 1-2 years from now (more on that in an upcoming article).

What It Means for Spending

When people are feeling uncertain about the economy and about their own place in it, they tend to pull back on spending, particularly on big-ticket items like home purchases and other significant life moves. And for better or worse, we are a consumer-driven economy. If people pull back on spending, it will have negative ripple effects through everything.

What does this look like today? Look at the housing market. The average number of existing homes sold nationwide over the past decade has been about 5.2 million per year. But in both 2024 and 2025? The number was just over 4.0 million, a notably low amount by both recent and historical standards. Those years mark the lowest levels of home sales since the mid-1990s, made all the more notable in that the population is higher today than it was back then, so on a nominal basis, there should be many more sales today than there were back then.

A large portion of the decline in new home sales can be attributed to higher interest rates, especially in 2024. But what is all the more notable about the low home sale totals in 2025 is that conditions for homebuyers have actually improved over the past year: interest rates are lower, prices have eased off and even dropped in some markets, and there are greater options and less competition in the market. Buyers should be coming back to the market now and not staying on the sidelines, at least based purely on quantitative conditions out there.

The National Association of Realtors is certainly hoping so, having flagged in their monthly report this past week that existing home sales in January were down 8.8% from December, which is a notable drop even in this cold time of year. Home sales were down 4.4% as compared to January 2025. This comes even as prices have stayed relatively flat (up just 0.9% nationwide year-over-year). The Realtors noted that wages are now outpacing home price increases by a somewhat healthy margin, further evidence that buyers should be coming back into the market, and yet they are not, at least not yet. We will see if they start to this spring as the weather and, perhaps, the housing market thaw.

---

Ben Sprague lives and works in Bangor, Maine as a Senior V.P./Commercial Lending Officer for Damariscotta-based First National Bank. He previously worked as an investment advisor and graduated from Harvard University in 2006. Ben can be reached at ben.sprague@thefirst.com or bsprague1@gmail.com. Thoughts and opinions here do not represent First National Bank.

---

Weekly Round-Up

Here are a few things that caught my eye this week:

• The New York Times had a podcast about prediction markets (last week’s topic here at The Sunday Morning Post) that provided a great overview. Listen to it here or wherever you get your podcasts.

• Aziz Sunderji wrote a piece for his Home Economics Substack series about condos losing value relative to single-family homes. The reasons? Harder financing terms and rising insurance costs, among others. You can read “The Condo Crisis” here.

• Homebuilders in South Texas are asking Trump to lighten up on immigration enforcement at construction sites. A key line from the article in Politico:

  Construction executives have held multiple meetings over the last month with the White House and Congress to discuss how immigration busts on job sites and in communities are scaring away employees, making it more expensive to build homes in a market desperate for new supply. Beyond the affordability issue, the executives made an electability argument, raising concerns to GOP leaders that support among Hispanic voters is eroding, particularly in regions that swung to Trump in 2024.

  Representative Henry Cuellar (D-TX) reported out of his meetings with the South Texas Builders Association, “They started off with, ‘hey, we were all Trump supporters, and we thought he was going to secure the border and then kick out criminals, we just never thought that they were going to be coming after our folks, our workers, on that.’”

You can read my prediction article of this reality that I wrote one year ago, when I said:

The impact of mass deportations of immigrants, not to mention the anticipated declines in new migrants coming to the United States, will be devastating to the home construction market. I don’t care what your politics are on immigration, illegal or otherwise; the home construction market depends on immigrant labor. This is a fact…

…Now comes the part where politics meets policy, which then intersects back with political reality. There are many wealthy donors, for example, particularly on the Republican side, who own real estate firms, construction businesses, and farms (which, as noted above, are also specifically and significantly dependent on immigrant labor). Where will their support of Trump and his hard line on certain topics like immigration run afoul of their own bottom lines? It’s worth following, and is likely to lead to political and economic fireworks in the months ahead.

And so it has come to pass.