# AI Infrastructure Costs and Labor Market Shifts

**Podcast:** Marketplace
**Published:** 2026-01-30

## Transcript

Your electric bill is going up, but how much are AI data centers really to blame?
Plus, could lowballing yourself help you get a job?
From American Public Media, this is Marketplace.
In Washington, I'm Kimberly Adams in for Kyres Doll.
It's Thursday, January the 29th.
Good to have you along.
If you've noticed that your electric bill has been creeping up these past few years, you're in very good company.
According to the nonprofit Power Lines, electric and gas utilities asked for permission to increase rates by $31 billion last year.
That's double what they asked for the year before, and in most cases, those rate increases were approved.
Marketplace's Sabri Benochore takes a look at why this is happening.
Osiris Bali lives in Little Rock, Arkansas, and his electric and gas bills are about to explode again.
We're all experiencing this winter storm, and you have no choice but to run the heat all day long, especially if you have children.
Completely aside from the storm, his bills have already been rising.
But this went up, you know, a few hundred dollars over the last few years for gas and electricity here.
He is annoyed that the Public Service Commission keeps approving the utilities rate increases.
A lot of corporate welfare, I'll say.
Every utility market is different, but overall, electricity and natural gas are among the biggest drivers of inflation, according to nonprofit power lines.
Charles Hua is executive director.
Utility bills have gone up about 40% over the last five years.
It's not just because of data centers, which are geographically concentrated.
In fact, in a lot of places, it's not even mostly because of data centers.
Our grid is getting old.
We're not using it efficiently, so it costs a lot of money just to replace and repair and modernize our grid infrastructure.
For years, there was underinvestment in the power grid as a way to keep prices down, according to Thomas Rowlands Reese, head of power market analysis at Bloomberg.
Underinvestment in the grid hasn't been a problem because U.S.
power demand has been stagnant.
U.S.
power demand is definitely not stagnant anymore.
And the long-delayed upgrade bill is catching up with all of us, he says.
Dan Pickering is co-founder of Pickering Energy Partners.
Demand growth a decade ago was running at about half a percent per year.
In the last three or four years, it's ticked up to around one and a half percent per year.
Just from population growth, reshoring of industrial activity, and just more things that are electric, like cars.
Add on top of that, data centers, and demand growth is expected to increase from one and a half to two and a half percent each year.
But there's another thing driving electricity and gas costs up.
The market price of natural gas, which powers a lot of power plants.
One reason is because new shipping terminals are allowing us to export more of it, leaving less gas here at home and pushing up prices.
In New York, I'm Sabrie Beneshour for Marketplace.
Wall Street today, not a great day for some tech stocks, just to put it mildly.
We'll have the details when we do the numbers.
Those shares are up more than 50% over the last year, thanks in large part to artificial intelligence.
Marketplace's Daniel Ackerman explains.
Yes, Caterpillar builds excavators and dump trucks, says Kristen Owen, an analyst with Oppenheimer, but there's one component it makes for them that's driving the boom.
When you think about what helps those big machines that we know Caterpillar for, it's large engines.
And lately, the firm is connecting more and more of its engines to electrical generators.
Because there's one industry demanding a lot of them these days.
AI data centers.
We have on order over 600 natural gas reciprocating engines from Caterpillar.
Connor Andrus is with Jewel Power, which develops data centers.
We're working on a 4,000-acre campus in central Utah.
The project won't be drawing power from the grid.
Instead, it'll generate electricity on site with all those caterpillar engines.
Andrus says they're a good solution for data centers like his because they run on natural gas.
It's abundant.
And quickly is the operative term here, says David Victor, a professor of public policy at UC San Diego.
So the folks who are developing these AI data centers are in the middle of an arms race.
And power supply is often the bottleneck.
You're raising massive amounts of capital and trying to deploy that as quickly as possible.
Time to market for your power projects really matters.
Connecting to the grid can take years of permitting and reviews, so many data centers are opting to power themselves with solar or wind, or even by reopening nuclear plants.
Connor Andris says he considered solar and geothermal for his data center in Utah, but the fact of the matter is those are hard to scale, an analyst with Melius Research.
So if you want power before that, and you're not already on the list, your options have to change.
Caterpillar says right now power generation is its fastest growing segment.
I'm Daniel Ackerman for Marketplace.
Turning now to yet another story tied to the tech world: a series of lawsuits filed against some of the biggest social media companies: Meta, Snap, YouTube, TikTok.
There are a bunch of cases, one of which started jury selection this week.
And these lawsuits aren't about the content that's on social media platforms, but rather about the nature of the platforms themselves and whether they're designed to get young users addicted.
Here to talk about what's at stake is Cecilia Kong, a tech reporter at the New York Times.
Cecilia, welcome back to the show.
Thanks for having me again.
So lay out for me this particular case at hand here and then why the plaintiffs think that their arguments are going to win this time.
Yeah, so in this first sort of set or tranche of cases in a California Superior Court in Los Angeles, we are going to hear from nine individuals who are suing all four companies.
And in this first case, Snap and TikTok settled with the plaintiff who is going by the initials, KGM.
What we know about her is that she is now 20 years old, but she began using YouTube in elementary school.
And soon after picked up Snap, and then also subscribed to Musicly, which is now known as TikTok, and then Instagram.
And her claims are that she just could not stop using the different apps.
And she became really hooked, particularly by algorithms that directed her to more and more enticing content that also happened to be quite toxic, leading to trouble sleeping, body dysmorphia and body image issues, and thoughts of self-harm.
And so she's focusing on, and all these cases are they're focusing on the technology itself, the tools and the features that are used to get users and particularly young people to engage more.
Things like infinite scroll, beauty filters that basically change your appearance on your face.
And so you you look almost cosmetically altered, actually, as well as like autoplay of videos, where on YouTube, I'm sure you've experienced this, where right after you watch a video, another one just starts.
Another one comes up.
That's right.
And you didn't even ask for it, right?
And this is all done through their algorithms of recommendation.
And these are not by accident, any of these features.
They're all done to get people to want to come back every day and often and for long periods of time.
What happens if these plaintiffs win?
So, what these plaintiffs are asking for are monetary damages.
And in a whole second set of cases that'll begin the summer, school districts and state attorneys generals are also suing, and they're looking for really big changes within these platforms.
They want Instagram to stop recommending content to young people that could be toxic.
They want the autoplay feature of videos on YouTube to stop.
They want beauty filters to be banned.
So they want structural changes that they will do through what's known as injunctive relief if they win.
You highlight in your reporting that some people are calling this social media's big tobacco moment.
Why is that?
Do you think that comparison is apt?
And how might it change the way that we even talk about social media and social media addiction in this country?
So I think it's very fair in terms of the legal strategy here in that what happened with the big tobacco trials was that over several years, attorneys generals and individuals sued the biggest tobacco companies for creating addictive products that were marketed to young people and for hiding the information that they knew internally about how harmful their products are.
The exact same thing is happening in these trials.
What the plaintiffs in these trials are saying is that these companies have studied and have acknowledged and have known about the potential harms of their products for quite some time, and that the CEOs either ignored the advice and/or the companies decided that it was too risky to the business model to change any of these technologies because they were so concerned about engagement going down and traffic going down.
And so that's very similar to what was claimed in the tobacco trials, in that they were personal injury claims and there was a failure to warn legal argument that the companies knew what was happening and they didn't tell the public.
A whole generation of young people have struggled with regulating their own use of social media, but the fault and the finger pointing being toward the young person.
And I think what this will do is turn the tables so that people will interrogate more the actions of the companies.
And that'll be a relief in some ways and a change at least in the way that society thinks about how maybe children in some ways have been the guinea pigs and maybe even the victims of businesses that have not put them first.
At least that's what's being alleged in these cases.
Cecilia Kong covers technology and regulatory policy for the New York Times.
Thank you so much.
Thank you.
Big tech firms diverged today in the markets.
Microsoft corporations sank 10% with concerns that they're spending too much on AI.
Those concerns didn't stop meta platforms from soaring 10%.
Amazon, which announced 16,000 job cuts yesterday, lost half a percent today.
Dan Ackerman was just talking about the heavy equipment building our AI future.
Caterpillar dug up three and four-tenths percent.
Rival Deer and Company, known for John Deere Tractors, added 1%.
Bonds rose, the yield on the 10-year T-Note fell to 4.23%, and you're listening to Marketplace.
This is Marketplace.
I'm Kimberly Adams.
It's no secret that the job market is tough right now, especially for new and recent college graduates.
A report from the Sen Gage group that came out at the end of last year found that only 30% of new grads they surveyed had landed a full-time job.
As the job market gets tougher, applicants are trying all sorts of strategies to get hired, including lowballing their own salary offers.
Megan Robinson is a freelance journalist who has a story in Slate titled, I offered to take less money to get hired.
It still didn't work.
Megan, thanks for joining us.
Thank you.
Before we get to your story, we should start with some definitions.
When you talk about low balling salary, what do you mean?
Um, low bowling is when job applicants offer an employer to work for less than the listed salary.
Candidates do it to appear cheaper and therefore hopefully more desirable to employers.
It's something I've done myself and have seen amongst other younger female job seekers.
But in my own case, and in that other women I interviewed, it didn't help any of us get a job.
I think it's more of a symptom of how difficult the job market is for young people right now.
I imagine that's going to be shocking to some of the folks listening that you would willingly offer to take less than the posted salary for a job.
Can you walk us through what led you to do that?
Yeah, absolutely.
So after graduating from my master's degree last year, I applied to over a thousand jobs and only had three interviews.
One of these interviews was for a writer-editor role at a small health publication.
But after my first interview, I was rejected.
It made me really panicky because I'd been unemployed for so long and I didn't know when I might get another interview.
So I was quite desperate, and I followed up saying I'd be willing to accept less than the posted salary if they'd keep me in consideration.
Surprisingly, it worked, and the hiring manager said he'd bring me back into the running.
But in the end, I still wasn't hired.
I was told that I wasn't meeting output expectations and they didn't have the budget or the capacity to train me.
It made me wonder whether anyone had done the same.
So I put a call out to my network to ask, and everyone who replied was a woman in their 20s.
They told me that they'd also offered to work for less, but in none of their cases had it led to a job offer.
You write that when you did put that sort of call out to your network and share your experience.
Some people actually chastised you for lowballing your salary, especially given the fact that women have historically made less than men in terms of salary.
What was your response to that?
Yeah, I mean, I understand it because on the one hand, it could bring down salary benchmarks for everyone, so I can see why some people were criticizing me.
I also can understand that like people were criticizing me because they said it's just a really bad um salary negotiation technique.
And that is also true.
I don't think employers view it very highly.
I think it signals a lack of self-esteem or a lack of understanding of the role.
And in my case, it didn't lead to a job offer, and in the people I interviewed it didn't lead to a job offer either.
Do you have any sense of how widespread this practice is?
Yeah, there's no official data on how widespread it is, but everyone I interviewed and the people who responded on Reddit were women.
And I think we see it more often among young women because women already enter the labor market with lower pay expectations than men.
Um a research study by Handshake found that Gen Z women expect to earn $6,200 less than their male peers.
Many companies have to post their salary ranges, but often these ranges reflect the lower end of what they've budgeted to pay.
So if then a candidate low balls that budget, they could be pricing themselves well below what a company has the budget to pay.
And if you're a woman well below other male applicants.
The pushback that a lot of folks often lay out when they hear about these sorts of stories, is that you know, choose a different career, pick a different career path, shift gears rather than continuing to try for the same thing that's clearly not working.
What are you doing now?
I'm lucky because I was able to sublet my apartment in New York and move in with a friend in Texas for free where I don't have to pay rent, and I'm focusing on freelance writing, and I would love to build that into a sustainable income stream.
And um I'm still applying for jobs, but I just apply for less jobs, so it doesn't take up as much as my of my energy, and hopefully something will come through eventually.
But um, one of the silver linings of this current job market is that it could push a lot of people to innovate out of necessity.
We are seeing a higher number of startups in the US and the UK, so hopefully that's a sign people are trying to build alternatives to traditional pathways.
But it's also very sad that a lot of um university graduates don't have a guarantee of work at the end of their degrees anymore.
I suppose what's your takeaway for younger workers still applying, especially given that your lowballing strategy didn't work.
I think a lot of job advice now is that it is a numbers game, but I would also encourage young people to do whatever they can to regain a sense of control over their lives.
For me, it's focusing on writing, but for other people it might be vocational training, and it's more satisfying to work hard at something where you have control over the outcome and isn't dependent on someone hang handing you an opportunity.
Well, good luck with the job search and the writing.
Megan Robinson is a freelance writer.
Her story is in slate.
Thank you so much for joining us.
Thank you so much.
One of the most common healthcare headaches is just how hard it can be to just get in to see a doctor.
According to a report from ECG management consultants, it takes on average a month to get an appointment.
But there's one model called direct primary care, where companies are trying to reduce those times by not taking insurance.
Instead, they charge patients a monthly fee for unlimited appointments, calls, and in-office care.
And starting this year, people can now use pre-tax dollars to pay for these memberships.
Alex Olgan has the story.
Dr.
Jeff Davenport was an early adopter of direct primary care.
He runs a membership-based practice, One Focus Medical, north of Oklahoma City.
I have been in this direct primary care practice since April of 2014.
He likes the pace of work, six patients a day instead of 25, and the flexibility.
Patients can text him for help with a sprained ankle or come in the same day for a sinus infection.
Davenport only takes cash, no insurance.
Concierge doctors often do both, charge the retainer and bill insurance.
He charges a flat monthly fee of between 70 and 100 dollars, depending on the patient's age.
Davenport has been advocating for the government to get rid of an old rule that blocked consumers from using pre-tax dollars for his monthly fees, which they considered insurance.
And last year, Congress finally changed it.
We've been fighting this fight for 15 years.
This is medical care.
It was really dumb that it was not fixed a decade ago, but the wheels of government move slowly.
Consumers can now use health savings accounts, or HSAs for these direct primary care fees, and that's opening up a big new market.
Forty million people had HSAs last year, and that could grow by another 4 million.
Rebecca Springer says employers are hungry for solutions to the massive problem of runaway health care costs.
She's with Bailey and Company, a healthcare-focused investment bank.
There's probably no hotter investment category in the private equity ecosystem right now than employer health care.
As in employer-sponsored health insurance.
And now direct primary care is appealing to investors too, because the subscription-based model is predictable, and cutting out the insurer can save money.
Plus, the regulatory change means millions of new potential patients.
Right now, most of the market is still made up of small independent practices with fewer than five clinicians, making it an attractive play for investors, says Springer.
Because there are established business models and sort of a long tail of emerging providers, it is a space that has a fair amount of greenfield uh opportunity for investors.
Health economist Dan Polski with Johns Hopkins University cautions against interpreting the increase in direct primary care as a panacea for the problems plaguing the nation's primary care system.
There just aren't enough primary care providers for direct primary care to be a national solution.
And because people with HSA's tend to be higher income, with the majority being people who make over $100,000 a year, this change does little for those who are already struggling the most to afford care.
I just don't want it to be a distraction.
From the larger problem, primary care is still expensive and hard to find for many people with insurance.
In Oklahoma, Dr.
Davenport is concerned the influx of money could change the model itself.
I do worry, though, a little bit, because of this occurring, there's probably going to be some venture capital trying to reach out there and tap into that market and create what some of us term DPC in name only clinics.
Davenport and other physicians were drawn to the model because of the independence and more time with patients.
He fears that investor money with its focus on growth and returns could undermine that.
This final note on the way out today, staying on the topic of healthcare.
We got some updated data this week about how folks are reacting to those expiring healthcare subsidies that triggered that record-breaking government shutdown.
Those enhanced subsidies did expire.
And now the health policy research organization, KFF, reports that more than a million fewer people signed up for plans through the Affordable Care Act marketplace during open enrollment this year compared to last year.
And that's just the drop in enrollment, which doesn't capture people who maybe downgraded their plans or are just finding some way to absorb the higher premiums.
Which, according to KFF, for folks who got the subsidies and wanted to stay in the same plan, jumped an average of 114% from last year to this year.
Our daily production team includes Livy Burdett, Andy Corbin, Maria Hollenhorst, Sarah Leeson, Sean McHenry, Michaela Sia, and Sophia Torenzio.
Will Story is the supervising senior producer, and I'm Kimberly Adams.
We'll see you tomorrow, everybody.
This is APM.
For second and third generation kids, holding on to your family's culture can be difficult and expensive.
I'm Rima Gheres, and this week on This Is Uncomfortable, I talk with author and journalist Aamon Ismail about how passing down his Egyptian roots to his kids has become a line item in his monthly budget.
I don't want to have kids who can't speak Arabic or read Arabic.
This is the only chance I get to do this, right?
When they're really, really young.
I won't get another chance to help guide them and like help them learn Arabic for the sake of them being connected to this massive heritage that they're inheriting.