Non-paywall:<p><a href="https://www.msn.com/en-us/money/other/us-is-starting-to-see-heavy-job-losses-in-roles-exposed-to-ai/ar-AA23iPFc" rel="nofollow">https://www.msn.com/en-us/money/other/us-is-starting-to-see-...</a>
Every interaction I have with AI where I previously would’ve interacted with a person is worse. Whether that’s something mundane like scheduling an appointment, use at a doctor’s office, chat support. I imagine it’s cheaper for the company deploying it, but it’s all bad. It lines up with companies being more concerned with investors than actual users and customers though.
The reality no one ever wants to talk about is that big tech doesn't need as many employees that it has. Anyone who has worked in big tech knows it is true.
For a long time this was compounded by a "better inside the tent pissing out than outside the tent pissing in" strategy, where you hired people to deny them to your competition and then invented things for them to do. If it turns out whole segments are in fact pointless then the endgame could be truly nuts.
This is very true and has been for a while. AI is just giving CEOs an excuse to just trim that fat.<p>Even after layoffs most of big tech could cut 50% of its workforce tomorrow and see minimal impact long term. Thats driving a lot of the present layoffs as CEOs go “oh crap we gotta look like we’re with the cool kids on this AI stuff”… and then just do a big cut and cite AI. Truth is they could have cut years ago but had no social pressure to do so.
I think Jensen Huang said this recently, and I've had a similar opinion for a while, but a lot of companies seem uncreative with how they use their employees. like google probably has >10k people working on stuff like "ensure gmail refresh button is the correct size", but why not fund teams to take on new and more risky projects... maybe part of it is that the type of people who work at big tech companies are not interested in risky projects :shrug:
Have you read the article? This is about professions like paralegals or translators, not engineers.
Okay but - if an industry can afford more employees than it has is there a real reason not to have them? Aside from shareholder greed etc.
Tech as a whole may not have too many employees, it's just Big Tech that has too many. If those employees were distributed among many more smaller companies the tech market as a whole would be much more competitive.
Yes because the most optimized company builds the most wealth and power and puts you out of business or destroys you.
Maybe a reasonable question in isolation, but "greed" is doing a lot of heavy lifting here. People's retirement accounts include shares in businesses because they expect those businesses to pursue profit and act in their interests as shareholders. If they didn't, people would rightfully pull their money and put it into more responsible businesses (or savings accounts, etc.). Charities and government-funded social programs exist too, but those are different things.
Adds more coordination…<p>IMO, rather than hiring more people than necessary, we should have a safety-net so that our society isn’t structured around “work or die.” I bet we’d see much lower admin costs at that point, as people aren’t incentivized to find less-essential tasks.
One reason would be that if they can create more value in another industry
This has been true for many years though
Almost like the whole “learn how to code” thing was a play. Commoditize your complement and all.
h1b and outsourcing will also go down on the tech side.<p>the group got hit the hardest is gen-z tech graduates, no fixes ahead for them,sigh
> A group of 18 occupations flagged by the Bureau of Labor Statistics as exposed to AI, accounting for about 10 million jobs, saw a 0.2% drop in employment between May 2024 and May 2025, according to annual data published Friday. That compared with an increase in overall employment of 0.8% over the same period.<p><a href="https://www.bls.gov/opub/mlr/2024/article/industry-and-occupational-employment-projections-overview-and-highlights-2023-33.htm" rel="nofollow">https://www.bls.gov/opub/mlr/2024/article/industry-and-occup...</a><p><a href="https://www.bls.gov/news.release/ocwage.htm" rel="nofollow">https://www.bls.gov/news.release/ocwage.htm</a><p>The article that flags AI risk still projects that the top 15 growth employment through 2033 contains software related jobs like Computer / Information Research Scientists, Data Scientists, Operations Research analysts.
I think people are blaming AI for a recession caused by trump’s tarrifs and the oil crisis. Businesses fear that oil prices may explode until enough of the economy enters a recession that oil demand decreases, and are already feeling the supply crunch
This comment is completely divorced from the facts given in the article.<p>The main point is that <i>certain types</i> of jobs like customer service reps, secretaries and sales people are being disproportionately affected. If it was just general fears about the economy overall one would expect a more broad-based impact.
I think it's more likely an interaction of both. Tariffs and recession lead to companies looking for ways to save money, and AI is both a convenient excuse for reducing headcount, and at least in some cases may actually reducing headcount without hurting productivity too much.
Salespeople feels more like a leading indicator of concerns about a recession than AI getting good. They're a profit centre, not a cost centre and AI automating prospecting some aspects of leadgen and the necessary but low value add stuff like meeting notes and formatting proposals should make them more productive not redundant, if anyone's buying. And AI is somewhere between very bad and completely incapable of most other aspects of their job (product demos, networking, in-person meetings, negotiating)<p>Especially when it's actually "services sales jobs a little up, manufacturing a little down"...
Those are the same roles that get axed first in any recession. Anything under “SG&A” is overhead that can be reduced.
If you replace customer service with LLMs I will look for a way to stop doing business with you.
The number of businesses you can switch to is quickly diminishing.
Competitors will be using LLM too.
Why care if "unthinking execution of a generic customer service script" is performed by an interpreter made of silicon or flesh?<p>Neither has any free will to deviate from the script. Both are useless in any case that's not handled by the script. The silicon one has better wait times though.
You will probably be in the minority, and that minority will keep shrinking as AI keeps improving.
Defending the sociopath overlords: truly a warrior's flex...<p>This one really triggers you guys doesn't it? What if I told you there are less than 1,500 billionaires in the country and you aren't ever going to be one of them?<p>You can do quite well but aligning with these f***wits isn't going to get you there because you need to provide something they don't yet know they need. And you won't get there and you won't build it if all you do is agree with them.
Yeah but the CEOs said it, so it must be true.
It's depressing to see logic and reasoning in discussion replaced with name-calling and tribalism.<p>"You're on the side of the <insert pejorative descriptive here>" should rarely (if ever) be a valid rebuttal.
Nothing they said is a defense of anything.
Are the jobs going away because of AI being applied or is AI being applied because those jobs are going away? You see lots of life jackets being worn as a ship is sinking; doesn't mean people putting on the life jackets made the ship sink.
In economies that revolve around credit/debt (the US), once credit stops <i>EXPANDING</i>, that's a huge force pushing the economy <i>down</i>.<p>If everyone's already taken on as much debt as they can, and interest rates aren't consistently marching lower for ~25 straight years, it's going to be a little harder to grow.<p>Doesn't make it impossible, just harder.<p>We're paying for growth we got 25 years ago now.<p>Wish we invested that money better. You can't change the past. But you can change the future.
This is precisely why DJT wants to control the Fed and lower interest rates. It would be a short-term win for him and the economy.
The future will certainly change things, but who can change the direction of a tidal wave? It will go where it wants, not where we want. The best a single individual can do is flee to higher ground.<p>Unless you happen to have a media empire at your disposal. Maybe then you can change the future. The problem is that those people are changing things for the worse.
The overall US economy is less impacted by higher oil prices than ever before. As a net oil exporter now in a way it's actually a positive, although the effects vary by sector.
We have an invention that aims to replace labor and employers want nothing more than to replace labor. It's silly to pretend that AI isn't responsible. But you're right that an incompetent administration is piling on with additional factors.
^ this!
And higher interest rates than before
Over hiring from Covid is still an overhang
GDP is not suffering — quite the opposite, 2% real annual growth in Q1 — so recession is definitionally inaccurate.
We’ll need new terms if Musk and Altman make infinity GDP and everyone else is unemployable.
More like Dario rather than Musk and Altman. Anthropic is much more likely to be a monopoly, and they are also pushing hard for regulatory capture with their whole false superiority on ethics, safety, morals etc.
Let's use percentage of people below the poverty line ( the actual poverty line is estimated between 97k(single) - 145k(family with children)<p>Anybody reading hackernews that cares about GDP growth is profoundly misinformed.<p><a href="https://www.cbsnews.com/news/income-needed-get-ahead-145k-half-americans-study/" rel="nofollow">https://www.cbsnews.com/news/income-needed-get-ahead-145k-ha...</a>
As a Brit, it seems hilarious that a single person with an income of $97k would be on the poverty line. That's more than double the median income for all households in the UK. A lot of Americans don't seem to realise how fantastically wealthy they are compared to the rest of the world.<p><a href="https://www.ons.gov.uk/peoplepopulationandcommunity/personalandhouseholdfinances/incomeandwealth/bulletins/householddisposableincomeandinequality/financialyearending2024" rel="nofollow">https://www.ons.gov.uk/peoplepopulationandcommunity/personal...</a>
Comparing countries can be relatively pointless because of all the differences. It only matters if you can obtain money in one country and spend it in another.<p>The main way these numbers are gamed is by finagling with healthcare and post-secondary education, each of which can be as high or as low as you want, depending on what you’re torturing the numbers to confess.
It's more than most Americans make by a large margin.
That’s a completely absurd number. No single person making just under $100k a year is living in poverty. Even in a high cost of living area.
$97k per year isn’t that much when health insurance premiums are $500 to $1,500 per person per month with $5k to $15k annual out of pocket maximums. You have to build up decent savings to ensure you can take care of yourself for when you lose your income.
2% is not the opposite of suffering. It's suffering-adjacent. It loses ground to China.
The US GDP per capita is still well ahead of China. The fact that China is catching up is not a problem for the US.<p>On its own terms, 2% isn't great. It's kind of the bare minimum. Anything less is losing ground in absolute terms, since that's the target for inflation as well.<p>That's also assuming the number is trustworthy. The number is a mediocre metric to start with, and I don't trust the administration to be calculating it correctly. They have shot the messenger more than once, after which the messages started to improve with no change in policy.<p>I would also not that China's numbers are even more opaque and less reliable.
Real GDP growth in Q4 was 0.5%, so it’s a heck of a lot better than that.<p>2% is average for past decade+.
GDP can be misleading. Lots of things are double counted.
Yet the stock market is reaching ATH, I don't think recession is the accurate word. I also think the fixation on Trump is misleading and takes away from the actual structural side effects of dollar issuance via unbacked fiat. It's convenient scapegoating to take your attention away from the long lingering issue of US dollar dominance and debt that is now causing large social divide and standard of living.<p>The physics of a reserve currency is that it reverts to zero through endless supply to fuel the engine to accelerate inequality at home and abroad. You see that an increasingly smaller club of people are politically opposed yet benefit by creating a narrative that takes your attention away from it.<p>This is the uncomfortable truth that not enough people understand and they are stuck in a loop chasing after things invented and created specifically to keep you there.
It shouldn't require pointing out, but asset prices are not a measure of economic activity. Rain amount and air temperature aren't either, if somehow somebody needs that pointed too. Those are independent things.<p>You are describing stagflation, that is a kind of recession. Despite all the fundamentals pointing towards it for a while, the US hasn't seen a lot of inflation (at least yet), so that's not what is happening.
The stock market isn't the economy. Note that stock prices are denominated in dollars. If the dollar price of a stock doubles, but in the same time period the value of the dollar halves, then the stock's value hasn't gone up at all. Compare the price of the Dow Jones to to gold and its value peaked in 1999, and has been nosediving for the past 18 months[0]. The ratio right now is the same as it was in late 2008. Obviously gold isn't a perfect standard candle, but given that inflation has obviously been rampant in recent years, we should obviously treat any economic metric that does not control for inflation with healthy skepticism.<p>[0] <a href="https://www.macrotrends.net/1378/dow-to-gold-ratio-100-year-historical-chart" rel="nofollow">https://www.macrotrends.net/1378/dow-to-gold-ratio-100-year-...</a>
Note for others - "ATH" => "All-Time High"
Well said.<p>Although the stock market isn't a reflection of the economy, the most relevant concern is the staggering 40TN+ of debt that can't ever go down as long as the dollar is the reserve currency and have no choice but to endlessly supply more dollars.<p>The issue with oil rising will make it completely impossible for the Federal Reserve to cut any rates and instead will either hold or raise them.<p>Then the stock market will have a problem; and those that have properties tied to their RSUs will start panicking.
<i>This is the uncomfortable truth that not enough people understand and they are stuck in a loop chasing after things invented and created specifically to keep you there.</i><p>Hardly. Those 'invented things' are invented for one purpose, to make the inventor money. That's how capitalism works.<p>Fun fact, let's say that tomorrow, all new credit was outlawed. And by law, all existing credit would require payments that took 3 years to pay off. In this fantasy world where we ignore mortgages, my point is that salaries would diminish.<p>Right now, the cost of everything a person buys is factored into salary costs. Take away interest, and all those 30% interest credit card debts would eventually no longer be part of your salary. This process takes years of course, but no one would issue credit cards, if every person defaulted on them, yet if you look at where salaries really go? With a lot of people, 30%+ goes to just holding debt.
> Yet the stock market is reaching ATH, I don't think recession is the accurate word.<p>It did under Biden as well yet nobody had any issues claiming the economy was in terrible shape and that we were entering a potential recession. Which frankly I agreed with! The stock market tells a fraction of the story. Especially because institutional investors now dominate investment in a way they didn’t only a few decades ago. A booming stock market does not benefit “the common man” like it once did unless one believes in trickle down economics or something. Employee pensions have become rare in the private sector, matched 401(k)s used for investment are also becoming more scarce. The stock market just isn’t tied to “the average person” like it once was, relatively speaking.<p>I do think Biden deserves immense blame for inflation getting out of control, but I also think as usual people overlooked what the Democrat president was handed, which was a Covid economy in this case. There was going to be fallout from that for years no matter what he did. No one blames Trump for the economy he handed off either.
That thing that reassures me is that I kind of think most people in software aren't doing much useful work anyway, so if the whole thing was really driven by hiring the fewest humans possible to get the work done, companies probably could have fired half of us even without AI.
I think the bigger issue is that management clearly have no idea how to tell which people are actually working. Output could be higher with less people but the situation that managers have crafted is the maximum number of people with minimum level of output. Not good.<p>The answer to every problem at my place is: more headcount, productivity drops further and further, more headcount, more headcount.
Because that's the incentive that management faces. Their promotion is dependent upon having more headcount under them. The key metric in their resume that determines which jobs they are qualified for is how many people did they manage. They don't personally pay for that headcount. If they meet some baseline of output people don't really ask questions (or are able to judge) whether that headcount was necessary. So of course they seek more headcount.<p>I suspect the economy would look very different if total headcount in a manager's org was the <i>denominator</i> in a manager's performance review, such that if you employ 10x as many people, you better have generated 10x as much profit. But this would also have lots of unintended consequences: management would be incentivized to employ as few people as possible, which means lots of people would be out of work and would be competing with your firm.
This matches my experience at MSFT. Very little work for too many people, MSFT CFO also said in the latest earnings calls that MSFT will have fewer people next year than this year, they offered buyouts to tenured people, did layoffs in LinkedIn not sure if there’s more to come.
I agree that a lot of people aren’t working at 100% and a lot of the work was probably needless busywork.<p>I don’t think it’s reassuring the way this goes away from AI layoffs. Just like in life, you need downtime and boredom. This is when you learn and reflect and develop taste, this is where your team steals time when an “emergency” breaks out and someone gets paged over an outage. This is where you take time to mentor the younger employees.<p>Relatedly, it seems that AI is entrenching cumbersome processes because it’s easy to write that extra doc, compile those notes, etc. When you make a team more lean, you’re supposed to cut out the extra process and let people work in a high-trust environment. That’s why startups are fast - everything is low process and the business has to trust employees. As companies grow, they develop all sorts of bureaucracy as scar tissue. But layoffs and cuts fail if you keep that stuff along the way.
This is a conspiracy theory, but I suspect that VC-backed tech industry overhiring circa 2020/2021 was partially done to give this impression.
There’s very little evidence of AI replacing jobs en mass.<p>There a ton of evidence that underperforming companies are trimming fat, with CEOs citing AI impacts rather than their own bad decisions that led to the fat that needed trimming.
AI is perfect for this because you get to sweep bad decisions you made under the rug and look like a forward-thinking visionary. Why would any CEO <i>not</i> blame their layoffs on AI, true or not?
Underperforming? Big tech companies are reporting record profits and major layoffs in the same press releases.
Losses in job categories management can claim was related to AI.<p>Some management may even believe AI can <i>fully</i> replace all work done by those positions. In some of those cases they will learn they were mistaken. The eventual rehiring of some positions will happen unevenly over time, be attributed by management to "ongoing growth" and not be reported by the media.
Looks like this data only goes until May 2025, so it's a year out of date. Also, overall employment is still up 0.8%
If its because of AI we should see a similar trend across other developed countries, no? And maybe even a correlation with how much they adopt AI <a href="https://www.visualcapitalist.com/ai-adoption-rates-by-country/" rel="nofollow">https://www.visualcapitalist.com/ai-adoption-rates-by-countr...</a>
Most of the job losses are in tech and it all started back in late 2022 and early 2023 arguably before AI was perceived as a real threat. I think after 3 years of tech hiring melting because of all the economic uncertainty people can't see the forest for the trees anymore. Justy $0.02
Here's a real-time layoffs tracker for anyone interested: <a href="https://jobs.voxos.ai/us/layoffs" rel="nofollow">https://jobs.voxos.ai/us/layoffs</a>
Honey, I overhired during the pandemic and I need a scapegoat for the layoffs!
Thats been mostly straightened out for years. This is tariffs, oil, and uncertainty more than anything.<p>When I worked in finance the thing that made the analysts happiest was a split election wherein neither side could accomplish much. We now have the opposite of that, and worse a capricious leader who makes huge changes without much justification or explanation. Pressing "shuffle" on the economic policy doesn't help anyone.<p>To finance the best government is the least effective government.
Fun fact the US government was designed to be ineffective on purpose. It's not supposed to do much <i>by design</i>
you just keep right on believing that but I'm on the front lines and we still have plenty of people that need to be laid off and we need a reason so AI is as good a reason as any other.
Yeah, the more data that comes out on the "AI employment disruption" the more it becomes apparent that AI is just an investor-friendly excuse to do the layoffs they were going to do anyway.
It is an <i>executive friendly excuse</i>.<p>No executive will ever want to say they made poor decisions that resulted in overhiring.<p>Instead, someone else's fault.<p>The same way for awhile the failings of certain businesses "those damn millennials refusing to spend money here" instead of "we priced ourselves out of the market".
Pandemic was 4-5 years ago. That’s no longer an excuse.
This is intentional. Companies are actively trying to develop capabilities to replace humans. I am surprised there are not enough protests against this.
Instead of protesting they are all getting together on Monday mornings in sf and hiking
why should I protest not needing to work for money?
Isn't post scarcity a great goal to have?
I am happy to see them replace humans with AI machines. However, housing, healthcare, and other insurances have become super expensive. Who is going to buy all the stuff these companies produce? Maybe, it will become a two tier economy: wealthy selling stuff to fellow wealthy people; under ground economy with bartering their trades for food, building shacks for living, flea market dentists, old Wild West style flea market medical doctors selling potions, elixirs, dry lizard oil, etc.
Exactly, the “economy” only matters because the ruling classes need the labor. If you could accomplish everything using machines you could have all the resources, beautiful vistas and global warming runway to yourself. Maybe some classical occupations exist like teachers, artisans, actors, sex workers so that the ruling class can still flex on each other.
i tried to protest forced-ai use at my company on the basis of it being a workplace hazard and was promptly emailed by their lawyers
Would like to see the numbers if we exclude healthcare jobs. I think job growth has been net-negative across the board when healthcare is excluded.
Whatever happened to the "AI won't replace you; someone who uses AI will" narrative?
Has anyone done the math on how many jobs AI has to replace in order to justify current valuations? Would be interesting to see imo
There's no public valuation of any AI company.<p>But by the investment contracts OpenAI announced, if they replaced every white collar job in the world and captured all of their salary, they would have a P/E of ~1/70. (Assuming no costs at all.)<p>You decide what return horizon you are comfortable with and solve the linear equation for it. (But my numbers are from the end of last year, I haven't seen any more recent compilation of their contracts.)
I think the goal is to replace workers who cost $10,000/mo with an AI agent that costs $8,000/mo
My back of the napkin was a floor of five to ten percent just to break even with the physical buildout + a couple of years of energy bills.
Some might be interested in the Iceberg Index: <a href="https://iceberg.mit.edu/report.pdf" rel="nofollow">https://iceberg.mit.edu/report.pdf</a><p>Basically, AI is replacing tasks of jobs, not jobs <i>per se</i>. There's a study by folks from MIT on what tasks, and which roles are most at risk.
It is quite clear that AI is only as useful as the person piloting it and therefore, if the job losses are due to AI, they are temporary. The C Suite folks spreading the batshit idea that AI can replace humans on LinkedIn will eventually lemming their way back to the correct side of reality.<p>I personally think that AI is being used as a convenient blanket to hide the plumes of smoke emanating from an economy engulfed in flames.
Not to worry. I've been assured many, many times that automation only ever creates more, better and higher paying jobs than it removes, and that no one but talentless bottom-feeding hacks are losing jobs to AI anyway, not the elite crème de la crème of HN. So the good times are on their way gentlemen!
> saw a 0.2% drop in employment between May 2024 and May 2025<p>So, the "AI job losses" just happen to be <i>before</i> AI started to boom, <i>during</i> the kickoff of largest trade war in 100 years?<p>They don't mention the trade war <i>at all</i> in the piece? And no analysis whatsoever of what other factors could have resulted in "customer service representatives and certain types of secretaries and salespeople" being let go. Who the hell writes this crap? What was the point?
Am I meant to pretend that companies weren't for years beforehand using RTO to drive layoffs? They are always doing it, AI is just the cover.
This is going to be horribly embarrassing when it turns out that AI can only outcompete human labor in some roles. <a href="https://www.axios.com/2026/04/26/ai-cost-human-workers" rel="nofollow">https://www.axios.com/2026/04/26/ai-cost-human-workers</a>
"Follow me: a day in the life of a soup line influencer"
Will the people wake up?<p>We need a modern take of the movie "They Live". Granted, any copy would not be anywhere near as good as the original, even for a B movie, but AI is really turning into skynet (but it sucks).
If you're checking on open-registration internet forums (HackerNews in particular), you will never see "the people" wake up.<p>If you're arguing against "AI" on HackerNews, you might as well join the Washington Generals. You and your opinion are tolerated only as long as there is a more popular pro-AI opinion in the replies to your comment.
<i>They Live</i> was a product of its time. We live in our time, so instead we've got <i>Sorry to Bother You</i>.
I'm not sure what 'film' America is trying to be but I've seen plenty of real life references that could have come straight from either the Handmaids Tale, Idiocracy, The Dead Zone or Civil War the past few years. Life does seem to imitate art sometimes.
No, the art is calling out what’s always been there but you didn’t notice.
Plenty of Do Not Look Up in there as well on a lot of topics.
It does not matter at this stage is an excuse for layoffs or not. The entire world should impose sanctions on the US for the largest IP theft in history until the laundering companies are illegal (and the waterways closed due to US wars are open).<p>If this administration works against its people and the whole world, maybe there should be a regime change. Peaceful version of Jan 6th. All the unemployed have little to lose.
Why just US and not China, France and UK? They all trained on internet text.
Let the entire world impose sanctions on China first if they care about intellectual theft
This <i>is</i> "AGI".
Discussions about this topic need to separate "AI" and aAI.<p>- "AI" is a tool which workers use - usually to produce outputs of the same or lower quality faster. It needs a human to be constantly monitoring, steering and verifying it.<p>- aAI is actual AI - what sci-fi authors imagined and what ML companies promise. You ask aAI a question and it either gives you a complete, unbiased and correct answer or says why it cannot do that (insufficient information, not enough computing power, etc.). It doesn't ignore instructions or get stuck in a loop. It doesn't delete your database or exfiltrate secrets.<p>Obviously, this is a spectrum, not a hard division.<p>Crucially, "AI" is a tool for workers. aAI is a tool for owners - it replaces workers. Owners give aAI control over a company or some other form of capital and it monitors, steers and verifies humans, who will still be economically valuable for manual work, at least until robots catch up.<p>---<p>All the people hyping AI either still see it as a tool ("AI", not aAI), not believing it will mature into aAI and replace them; or they are the owner class who will benefit from not having to pay wages which maintaining their passive income.<p>If most of your income comes from work (as opposed to passive income from ownership), you have 0 reason to be excited by AI. Your life will not be better if you lose your economic value.
The demand for cushy white collar desk jobs (which weren't around 30 years ago) is crashing. Rest of the world, not so much.
30 years ago was the middle of the .com boom and white collar jobs had been around for a century already. White collar employment started with the industrial revolution and really took off during the post-WWII boom.
Cushy white collar desk jobs were not around in the mid 90s?
Figure has been doing a 75 hour live stream of it actually replacing a real job and it has that AI magic where it has gestures and movements that seem emergent. They’re coming for anything manual in less than 2 years.