IT Unemployment Rises to 5.7% as AI Hits Tech Jobs
wsj.com
wsj.com
The article's evidence for the AI claim is just "some guy you've never heard of says so".
It works like this. Find whatever is hot that day, week or month then put it after the "as".
Stock market down as Trump holds another signing. Stock market up big as Musk slashes government spending.
* Something A (probably) happened
* Something B (probably) happened
* The author/publisher are slimy weasels and I'm not going to give their stupid article any more time
Also as you flatten organizations and use cloud services you can discard lots of people. Every large company had a storage team. I worked for a big company that probably had 250 FTEs just running storage. Not anymore.
There’s also a ton of uncertainty in the US. Everyone is banking on a significant correction.
I’d have like a quarter. Which isn’t a lot. But man, you’d think I’d have learned by now huh?
Also, man do I understand Ye Wenjie from [https://en.m.wikipedia.org/wiki/The_Three-Body_Problem_(nove...]. I don’t support, but holy shit do I understand.
Don’t get murdered pointlessly trying to do the right thing, eh? At least make it count for something useful. Try not to be dead right.
Reality is reality, and in the end you are right, but by that time it won't matter; the damage will have already been done. This is the nature of cascading failures, and delusion, and there are far more delusional people today in positions of power than ever before.
Edit: On a separate thing. As an FYI in case you didn't know, but it looks like your blog is now directing to a bad link on google, or they were hosting it and took your blog down.
They use wording -- just like this article does -- like "X happens as Y also happens".
The keyword being "as" and not "because" or "due to".
It implies causality, by coupling both in the same title, without technically linking the two.
E.g. "Stocks fall as rate increase looms"
They're basically just saying two things happened at the same time, or one after the other, but one didn't necessarily cause the other.
IOW, they know they can't link the two -- i.e. they can't user words like "because" or "due to" -- but they also know that by coupling them in the same headline, readers will still infer causality.
“President announces crackdown on immigration amid ongoing success of Taylor Swift’s new album”.
“My God, you’ve actually seen people looting, raping and eating each other?”
“No, we haven’t actually seen it Tom, we’re just reporting it.”
The accurate, but boring headline is almost always, "stock (tumble|gain) as algorithmic traders reinforce trend. Further news may occur."
I think I originally read that this happens in a book somewhere, then observed and noticed yeah that seems to be the norm. Not surprised it seems to be same thing in these adjacent metric % headlines where there's no proper thought if there exists a causal link between X and Y.
I think my brain has learned to recognize the pattern "Something rises X% as Thing Happened" and it reminds me of that cheeky quote about how every headline that is in the form of a question is, ... uh I forgot the full quote. But feels like almost this could have some kind of cheeky rule of its own, about how no causal link ever exists in a headline like this.
Black Swan by Nassim Taleb?
That book however you mentioned, I just looked it up and it seems interesting. Might put on my read list.
But specifically this is probably mostly driven so far by tech support industries. "IT" being customer service reps who use a computer to respond to chat messages, in some cases.
Or actual tech support that's had its entire entry system stripped and replaced by an AI-guided walkthrough of predefined steps customers are required to take before they talk to a human. This will certainly have impact on the amount of support staff needed, even if a portion of that is simply frustrating customers into searching for answers themselves.
How much tech support was still in the US though? Outside of small onshore teams, frontline tech support for the major players was already all in India/SEA. You could replace those people with AI, yes, but it wouldn't reflect in US unemployment stats.
They are replacing second tier people with people overseas who interact with chatbots and a limited number of staff in the US. So one dude can access customer data, but he’s servicing a dozen agents.
"Another reason for January’s tech job losses was that companies began implementing some intended spending cuts for this year, Janulaitis said, and many slashed budgets based on what the economy looked like during fiscal planning last year."
There are several important things to look at that show this. First and foremost, IT has about a 40 year history now, with corresponding economic data.
In its entire history, its been uncorrelated with interest rates and other assets. These correlated assets are what most people refer to when they speak of wider economic circumstance. The IT industry is uncorrelated with this, with 40 years to back that up.
IT labor markets are however strongly correlated with advances in information technology. This correlation is found throughout that dataset as well, in both the boom and bust cycles.
The deep unemployment which we are seeing here, which doesn't include the first batch of laid off employees two years ago [18 mo], is only correlated and impacted by Information Technology advances.
There is only one such major advance that has been made in this time period... Artificial Intelligence.
AI has improved dramatically from a modular swappable design, and it can replace entry level positional tasks completely. It probably won't ever be able to take over the senior level positions, or the mid level positions, but that doesn't matter.
The companies involved in integrating these things into the production environment have mislead most of their customers, and in the process burnt all the bridges. Many of these companies market towards replacing workers with AI, grossly negligent of the whipsaw they are creating with that misleading narrative. In the short term they make lots of profit, while destroying the environment they need to sustain themselves.
By choking off the professional development pipeline which is a sequential pipeline in developing talent, this leaves few opportunities to progress to mid and senior level positions, which were few and far between to begin with.
When there is no economic incentive to go into any specialized field because AI has burnt the only bridge for a long-term career, people don't go into these jobs. Worse, those caught surprised by this change in circumstance, who actually have sufficient experience to perform at these mid-level ranges, may abandon the profession causing exponential brain drain. As anyone knows, people age and eventually die. Senior level people are at greater risk to this than others.
This inevitably creates a tsunami of cost in those irreplaceable positions which cannot be addressed by the market, and balloons far above what the market can bare. The forward looking expectations have made any labor in this industry worth less than AI services, and IT as a general rule is a labor multiplier where changes here eventually expand and infect everywhere.
Business is not constrained in hiring people because many business receive operations funding upfront from their financial engineering which often comes from money printers. By the time they notice the problem, it will be too late to solve it, in many respects its already too late because you have people who are incredibly competent, near geniuses, and they see no future in the industry and are retraining ahead of the curve.
AI threatens society because it disrupts the core pillars of society, which is indirect but immeasurably important to sustained organization of labor and food production. Agriculture is dependent on something like 64 different intermediate producers to maintain production levels sufficient to feed people.
Failures back to pre-modern industrial technology would mean half the people alive today starve to death, no matter how much they work.
These are the issues which the article doesn't really touch on correctly but its WSJ so what do you expect. Their journalism has never really been up to snuff.
In our company we're not seeing any major impact on productivity yet and no jobs were lost. Yet.
It is right there, in the headline, sir. The economy is so good, like never before. /s
The US is expensive, so offshore. Offshore projects are more expensive because they never work, so come back to the US.
I've lived it twice now. We’re in the middle of my third cycle.
In 2006 the Indians would randomly cut out, couldn't be on video (phone only) and had major language barriers, and they would join in their evening.
Now in 2024 you're on their schedule. Believe me, those people are not staying up late for your 3pm call. Wake up at 5am every day to keep your job, you have 3-4 hours to work with them.
The churn is real and any offshoring hotbed ends up with a lot of competing between big companies for talent which then drives the prices up and motivates people to jump companies more.
Not true at all. We are in the same regime as the pre 2008 rate response. We are below the late 90s boom times.
So you’d need some other explanation other than bare rates.
You see, it’s of critical importance that Santa Clara valley retain the character of a quaint mid 1960s town of single level shops and tract houses. Where jimmy can ride his Big Wheel in the Cul De Sac.
But back then, the Engineering Manager or Product Manager would remain in the US.
Now even Engineering and Product leadership is being outsourced as well, which makes it easier to enforce standards.
But after the COVID layoffs happened, impacted H1B, EB1/2, and O1 visa holders returned to their home countries, and then got rehired by the same employers but for $60k-130k TC (depending on country).
A lot of mid-late career product and engineering leadership was laid of in 2020-21, and those guys have now founded the new generation of foreign offices.
I've been employing eastern European IT folks since the early 00's as I've founded/worked for remote only companies since the late 90's. We were ahead of the game here, and anyone we hired was not treated like a sweatshop worker. They were hired and promoted like any hire we made in the US and were simply called an employee. The talent level was same or better than the US talent level for 1/10th of the cost, especially at the mid-tier/career level.
The largest issue was timezones, but if you needed to staff a 24x7 highly skilled ops team - this was one effective way to get a great head start doing so. Plus it was relatively easy getting folks to work odd hours if you were paying double local salaries.
Since the pandemic things have entirely shifted. It used to be large companies/enterprises would offshore to sweatshop style outsourcing companies. They had a hard time competing with us since the work and responsibility simply did not attract the best candidates. We could hire directly out of local university pipelines and be extremely picky in only taking the top 1% of candidates due to this. After the pandemic this totally changed to these large companies now competing on salary at levels we no longer found competitive. We simply no longer enjoy the strategic advantage of having an untapped low-cost highly skilled labor pool in these areas any longer. Large companies have more or less adopted our model.
It's now not uncommon to have entire teams headed up by a highly competent senior VP or C level local leader who reports directly to US upper level management. They are included in all business decisions as anyone in the US would be at their level, and performance is similar to any US team.
I have always (and even written a few times on HN about the subject) thought highly compensated US engineers were being exceedingly short-sighted in their demands for WFO since this was always going to be the outcome in the end. Once you can hire someone in the midwest US, it's not really a huge leap to realizing you can do the same 2500 miles away in a low cost country. There are still tons of labor arbitrage opportunities left in the world and the story that there is no local technical talent in these places is utterly false. It's always been a management problem.
Over time it will even out - and it has quite a bit. Salaries are now maybe 1/3 to 1/2 US salaries at the same talent level, so it still has a long way to go in my opinion. You are only seeing the start of many companies building up long-term strategic plans in these regions - these things take years to decades to be fully realized.
Most US WFH engineers will be competing on salary/skillsets in a world market within their lifetimes regardless of cost of living differences. The writing is on the wall, and I personally believe the die has been cast at this point.
In Cybersecurity and DevTooling we've moved the entire Engineering and Product function to Israel, India, and Eastern Europe (Czechia, Poland, Romania).
We'll still fund founders in the US, but they still end up hiring in those countries instead because you can pay "Austin in 2010" salaries and get top tier talent.
20 years ago, P/L responsibility would remain in the US, but in the current iteration, even P/L stakeholders are now abroad.
VPs have been explicit, in writing, that we are shifting heads to lower cost geos.
Hiring has been largely halted in the US for the bulk of non-business-critical roles to shift PCNs overseas, and there have been multiple waves of US focused layoffs to motivate this further.
While offshoring has been a topic for the last 20+ years of my career, this time, since 2022 or so, feels different even vs. prior recessions, and I expect we'll see continuing erosion in the quality of engineering positions and careers.
Additionally, can obviously only speak for myself but I've tended to see more hiring in countries without the strong european worker protections (eastern Europe, Israel, india, asia)
I feel bad in a way, considering that I probably stole some other engineer's job simply because I am cheaper.
It's a moral problem I hadn't(or rather, didn't want to) consider before reading this comment thread.
I love the people I work with(both overseas and local), but comments like these make me wonder if I've made the wrong choice.
> I love the people I work with(both overseas and local), but comments like these make me wonder if I've made the wrong choice
It's out of your control as an IC. That said, do whatever you can to prepare for the worst.
Your a competent engineer, just like anyone else in any other country.
I don't begrudge a romanian SWE any more than a US one, we're all just trying to get by here, and the fact that global megacorps can do currency/economy arbitrage that they prevent workers from benefiting from while ensuring they have every advantage is by and large not your fault.
It should have been for us as engineers to organize, unionize, or legislate protections if we didn't want this to happen, and despite advocating this to my peers during the boom years, I was constantly rebuffed.
Obviously I put more on the massive companies with the power to dictate global laws and trade practices gleefully jumping at this, but between that and the responsibility I put on my peers/myself to fight against that, I certainly don't put blame on you.
The only difference is that you have zero control over the weather and very nearly zero control over corporate hiring decisions.
Putin is covertly manipulating Poland with the same games as he did with Ukraine. And now it appears increasingly likely he has the right guy in the white house to take what he wants.
Great place for talent, pretty horrible for salaries and its going to get worse (from employer's perspective just to be clear, all the power to you guys there... I've lived 5 years in that city and if you are into European cities and city life in general and don't mind having mountains far this one is nice).
A lot of mid-level management on visas were laid off during 2020-21 during the depth of COVID, and this had to return.
Now, the SVP for a core Engineering Org or your PM Director owning an entire BU might be in Bangalore, Tel Aviv, or Cluj.
If ICs to mid/upper level leadership are in the foreign office, it doesn't make as much sense to keep hiring domestically for Product or Engineering roles.
Like, only now?! This had been the case my entire career to the point I got fed up of rat race to the delight of foreign superior, became cynical and unemployable. It's not ey-ay, nor outsourcing, it's the economay. We have a crisis, sir.
This was totally predictable. If your job can be shipped to Kansas from SV because it is cheaper, it won't be too long before it is shipped to Canada which cheaper than Kansas and then to Colombia.
But HN is still embracing remote work. I hope those in the US/Western Europe wake up soon otherwise all the coding jobs will go the way of manufacturing jobs very soon - either done by robots or done by cheaper labor in Asia or Africa or LatAm.
Small org, no low-performers, no layoffs. Same result as everywhere else.
From the article...
> "Jobs are being eliminated within the IT function which are routine and mundane, such as reporting, clerical administration,” Janulaitis said. “As they start looking at AI, they’re also looking at reducing the number of programmers, systems designers, hoping that AI is going to be able to provide them some value and have a good rate of return.”
Hoping... - I think this is where "leaders" are showcasing their ignorance. They're being sold a bill of goods that, truly, doesn't exist at the levels they think it does.
Now, on the flip side, the first part of the paragraph above seems plausible. However, these mundane tasks have been able to have been automated for a long time. It's just that it wasn't en vogue.
Reducing the roles outlined will truly lead to shittier outcomes for these companies and I hope the ones that are diving head first end up paying the price as it showcases, not their efficiencies but their ignorance.
AI has a few great use cases. Replacing IT folks who are skilled and can think on their feet are not a prime target for AI.
This still does replace workers in the end. It doesn't really matter if you lose your job to an autonomous AI or you lose your job because someone else is using an AI to do your job and theirs
I don't personally believe AI makes people this much more productive, but clearly people in charge of headcount are thinking it is
To add AI functionality, you will first need more developers, and a different kind (those that can train/evaluate/improve/deploy/integrate AI models) before potential savings can be kicking in. Some payback that may lead to layoffs could happen, but with a delay. It would be very silly to lay off a good developer rather than re-assign them, because the cost of hiring them is enormous, and you may not get the same quality again; in economic terminology, software engineers are "non-fungible resources".
and before you say this is false I've seen it firsthand many times. personally seen it with my eyes.
https://www.forbes.com/sites/jackkelly/2025/02/02/google-off...
https://www.forbes.com/sites/esatdedezade/2025/02/10/meta-jo...
And yet it’s causing massive disruption. The bosses have such a hardon for layoffs that they’re doing this on the slim chance that it might work for them. Having to hire programmers back in a panic is, as they see it, future someone else’s problem.
Yet.
If you give it a prompt which resembles a requirements document produced by a competent manager, you will get what you request and it will run first time.
No manager and coder pair can iterate on an idea 5 times per minute, even if the first attempt needs correction, the timescale disparity makes AI agents appealing for nearly all noncritical coding tasks.
It's just AI all the way down, and each step still needs to be validated because AI is not reliable
It isn’t a matter of iterating either, it’s being able to figure out what is politically workable and actually implementable.
No AI can solve that one.
Which the AI will just make a stab at without digging in and asking clarifying questions or examining all aspects of the codebase to spot what challenges there are with doing it that way with the specific custom business logic needs needed for the rest of the application, and then get it very wrong and not anywhere close to what was intended (although possibly exactly what was actually asked for).
Meanwhile even a slightly seasoned developer will know to ask for clarifying questions on various aspects of it, or point out challenges with the existing codebase to match the requested requirements (sometimes after taking a stab at it themselves and realizing it doesn't work).
There are ways to prompt the AI to produce better results, with very specific and detailed requirements that include tricky aspects of the technology (that developers are aware of and can provide when they write prompts), but the typical requirements that come down from management ain't anywhere near that.
Not saying AI won't ever get to that point, but we're not there yet.
I continue to hope that the net result of programmers getting more productive is an increase in demand for our skills, as employers realize they can get a whole lot more value for their investment. It might take a while for that to shake out though - we need companies to realize they can take on more ambitious projects now.
Let's say you're trying to build a product/feature. You have general timelines for the launch. About 5 years ago, it would take a team of 10 to deliver it. Now, let's assume there were productivity gains, and you can deliver it with a team of 8.
This would not necessarily be a problem if for 20% of the workforce cut, you would also start getting investments and create 20% more companies/features. The big uncertainty is, will this ever happen, or it'll end up with 10% increment in such activities, so you still have that extra person hanging around.
1) the headline is bait for a story content that's at best inconclusive, 2) of the two sources for this story, one analyst's point that IT spending on AI outpaced new hires seems to do most of the post's heavy lifting, 3) the second source -- an Indeed economist -- makes almost the opposite point that after increased spending on AI at the expense of new hires, spending on the is stabilizing 4) And this entire story appears to be based on *one month's worth of job reporting*
I understand why a WSJ reported is professionally required to spin something out of a nothingburger and convert attention into ad impressions and "provoke debate" or whatever, but for those of us uninterested in WSJ or Rupert Murdoch's success this feels as unhelpful as that post from a week or so ago about yet another (inconclusive A.F.) study on the effects of marijuanablahblahblah ...