AI isn't replacing jobs. AI spending is
fastcompany.com
fastcompany.com
All the while management is breathing down your neck and asking 'why isn't it ready yet'.
Once the thing is shipped, then all the important people come out of the woodwork, who were surely there all along, 'supporting' you from behind the scenes, there are photo ops and important people shaking hands. If they feel particularly charitable, then you might get a seat at the table. There's talk of spinning up a team around the product, and people fall over each other to get to lead it.
But the thing is, most likely they don't need your expertise any more, not really, once everything works, you don't really have a negotiating position as a dev. They get some cheap juniors to fix the bugs and add the missing feature niggles - hiring 3 juniors might not even be cheaper, the point is management does not have to depend on you, they can play their human resource games.
'But only I can fix that complex race condition, that popped up half a year after development' - well if it was good enough with the bug for people not to notice it for half a year, it's going to be good enough for another half, until the new devs can fix it.
This applies to ambitious feature requests as well - if the code's good enough that the contract was signed, the business requirement was met, they can just kick the can down the road until they can fix it.
Funny, this is the complete opposite of my experience. Greenfield projects I've been a part of have had a ton of highly visible progress with _frequent_ updates to stakeholders basically from day 1. Same goes for complex additional features.
Of course you can call out the former examples as incompetent or hubris, and they will probably occur less and less, but nevertheless they exist.
I think the OP meant projects that required long investments without immediate returns. For instance, a platform migration that requires many components to be finished before it can even be tested. Or, real greenfield, like a new product venture with a unproven customer thesis. The kind that requires months of work before you can go to market and validade... How do you report progress? Components built, percentage completed? Without something a user can drive or a seller sell, that's not progress... It's just speculation.
From my time at AWS, and in light of the recent DynamoDB dns race condition bug, this is a line of thinking that is problematic.
There is so much complexity in the inner workings of any one of those services. Many of those services are now composed of many teams, each having turned over completely multiples of times.
Yes there are runbooks and knowledge passed down from each generation, to some extent. But when you hit some critical outage, edge case, etc., those people running the systems have such a different relationship to the system having simply maintained it vs. built it.
Long stays in most companies are practically non-existent these days, apart from a few folks who are present for long, unfortunately these people are often called lifers, coasters etc.
Simply put, no one person can say they contributed most to the building of a system. Barring very few rare projects.
So you already need the runbook, documentation model to run things, for the reason people themselves work along those lines.
It's hard to commit to a green-field project that is predicated on a level of risk that leaders are hesitant to take on as they would also take on the "counterparty risk" of the expert individual contributors holding them over the barrel to finish the project. The expert individual contributors are likewise hesitant to devote themselves to the task knowing that once leadership considers the "hard bits" to be done, leadership's aversion to risk will try to swap out the expert-individual-contributor roles with much more replaceable roles and ultimately replace the expert individual contributors.
P.S. The parallels between this cycle of mistrust and the modern dating crisis (in the US) does not elude me.
They’ll add a few sleep(5) calls to make them go away though..
> They’ll add a few sleep(5) calls to make them go away though..
I don't think torginus's point is that the new devs will find the proper fixes for the code, more that such a hack might be good enough in the eyes of both the company's management and the company's users.
As much as it pains me to recognize this (as a fan of clean, elegant code) not every bit of software needs to be clean and elegant to achieve its intended purpose (which is, at least in the corporate software world being discussed here, to make money).
If you meet the needs of your software's users, you can make a lot of money for a lot of years selling a piece of crap held together with chewing gum and elastic bands (and many companies have).
You can build faster now that you ever have: I am building faster than I have in 25 years of engineering. You have more capable support for all the unfamiliar processes of building a business imaginable.
And almost everyone larger than you is finding it harder to achieve similar productivity gains from implementing AI, if not outright struggling with it. This is a golden moment and won't last long.
If anyone is worried about their job, it won't be AI that takes it - it will be outsourcing. The US offshores 300k jobs per year with a high percentage of them being IT (60-80% depending on source). It's really not that different to the offshoring of manufacturing decades ago. Why pay people onshore when you can pay someone in India half of that? Any job that doesn't require a physical presence or has legal pressures to keep it onshore will be at risk. It will likely get worse over time, just like manufacturing. I don't know what the future will look like if we continue outsourcing everything. It used to be that we outsource primary and secondary sector activies so that we could expand tertiary industries. What are we replacing the outsourced jobs with now?
We had British SaaS supplier and all the time I have talked with Bangalore-based people. They had to work in night I suppose…
Source: lived in India for a year.
It turns out there is little value in management alone. It will be cut loose like the rest. Perhaps these companies can become patent trolls.
Also they have a very hard time to find new talent since word gets around. Sometimes it works for industries like textile, but not always, and it doesn't work for engineering.
It mostly works for low budget or high volume products. It is still a bad long term strategy. Still, outsourcing didn't just start yesterday.
Even Japan has become pro-Indian on immigration for tech: https://eastasiaforum.org/2025/11/06/takaichis-japan-looks-t... Wtf...
So you basically can pay an Indian immigrant a junior dev salary, for significantly-better-than-junior-dev work. It's just stonks.
There’s a lot of opportunities for personal networks, nepotism and plain old corruption to work. (Who is going to figure out that somebody dropped a few gold coins to your sibling as a kickback?) There’s a much smaller network of people with relationships to Eastern European or other companies.
Here is an article putting together data from GitHub about SW developers per country.
https://data-player.com/highest-number-of-software-developer...
When talking about outsourcing jobs from the U.S., you’re obviously gonna exclude the U.S. China is also not a factor. Neither are Western European countries or countries like Japan and S Korea because of relatively high salaries. Russia is out due to long standing geopolitical issues.
That basically leaves Brazil and Indonesia as the only alternatives to India in the top 10 and combined they don’t even have half the number of SW developers as India.
You need to then add Mexico, Vietnam, Turkey, Philippines and Poland to the above 3 to add up to the number of SW developers present in India.
Thats why the outsourcing industry is concentrated in India. You can setup 1 office in India and have access to as many SW developers (Indians are also very willing to migrate domestically, so an office in a single city is sufficient to cater to the entire domestic developer market) as you would if you setup 8 offices in & different countries across 4 continents.
It’s genius, in a way.
They have a few very competent developers who are primarily in the secondary sales in my experience. First sales contact is between non-technical management and their front-line sales (usually very attractive women). In the second sales contact, technical staff from the potential client is involved and they bring along their real developers. But those are not the ones you'll get on the project. They'll give you interviews with the developers you can get, and they're coached for the interviews and sound fine. But then in reality they are people who can't touch type and develop purely by trial and error without forming a mental model.
If hiring locally wasn't such a mess, nobody would talk to them. At some point even a junior developer is better than not having a developer at all. I assume AI will change that and they'll get replaced first.
Indians are more visible, but salary expections have gotten extremely out of whack in the US, and extended WFH during COVID proved to most boards that companies can continue to operate when entire teams are communicating async.
If a large portion of interns and NCGs in the US are essentially expecting $45-70/hr salaries, it just isn't sustainable especially when factoring the growing skills deficit because universities failed CS students to a certain degree over the past 10 years by watering down programs in a short term bid to compete against bootcamps.
If we are paying Bay Area salaries, we expect performance comensurate to that salary. Basically, all companies are now starting to adopt the Netflix model of hiring in the US.
You mean cost of living and inflation has devalued the dollar.
I guess GCC night be Global Capability Center?
My other comment: https://news.ycombinator.com/item?id=45867014
With AI even dumb people feel smart - thanks to the overconfident answers, I am expecting this to be race to the bottom. Even the highly skilled offshore employee is not safe. Middle managers crunching numbers and looking at the bottom line might be convinced that if AI + less experienced but cheaper "human in the loop" can produce "similar" result at 10-20% of the original cost then that is worth it. I am personally dealing with such middle managers nowadays day in and day out. They seem blown away by Copilot demos and keep pushing "savings" and don't seem to understand AI hallucinations at all.
I think this is going to be a new golden age kf innovation. Lot of senior engineers capable of pencil whipping microservice infrastructure paired with industry experience means startup mill goes brrrr
Started doing automation about 6 years ago. We had several offshore teams. The one caveat was that the company was vying for several government contracts. One of the requirements was the ability to pass a low level security clearance and on yeah, you cannot, under any circumstances off shore any of the work.
Therefore, they needed an onshore team to handle the government work.
Well, when things got slow, our team was essentially the clean up team. We had several hockey players on our team so we nicknamed our team "Zamboni". A large chunk of what we were doing was fixing the off shore teams coding and things would break so regularly two of devs became full-time off shore support.
The company was literally paying our off-shore team to build an automation. Then when it constantly break, they would pull us in to fix it.
Absolutely maddening they couldn't understand they were paying triple or quadruple for work we could do once properly instead of two or three times before we got it.
The remainder of the talent tends to struggle with some of the outsourced work, but with AI they can now give a semblance of competence.
In the UK a major retailer, Mark and Spencer got hacked after outsourcing work to India. They couldn't fulfill orders online for months, and they are now reducing the amount of work they outsource to India.
We will see something similar happen to other companies in a year or two, but until then we just have to tighten our belts and hope we don't get layed off before then.
Just perusing on google for roles at google in hyderabad, I think its significantly lower than the US.
I've never heard of typical senior engineers getting paid $400k/yr in those locations.
Outside the US, I've only heard of Zurich being the only comparable location.
In those locations?
Based on sheer CS grad numbers why wouldn't companies just shift their r& operations there then?
One issue is that in many industries senior leadership just doesn't stick around for long enough. If your CEOs rotate out every 5-10 years then you're basically SOL; the next guy comes in, gets bamboozled by sweet talk of vastly reduced costs of offshoring and BOOM, round you go again!
Does sound more like correllation than causation. Was there evidence that the Indian devs made the mistakes that led to the hack or was it just the good old 'let's fall back on racism to avoid blame' by management? I still remember the articles where Boeing tried to peg the 737 MAX crashes on Indian engineers who worked for $10 an hour.
As other commenters have pointed out, this is simply bullshit. Good talent in India and the Philippines earn nowhere near US dev salaries. Unless by "very comparable" you mean 1/3 - 1/2.
Just outright insane.
1. It enables them to live in lower cost of living cities. This makes them more competitive relative to outsourcing because a lower wage in a cheaper city goes farther.
2. It opens up the job pool. If you work remote, you can work at any company that takes remote workers regardless of where you live.
3. It reduces the cost of switching jobs. Many people are stuck in jobs they don't want because there are few other local opportunities and switching jobs means uprooting and moving. For a single 20-something in an apartment, that doesn't sound so bad. But once you have a partner with their own career, kids with meaningful friendships, a mortgage, etc. then moving can be extremely disruptive.
In general, more job flexibility increases the efficiency of the job market for employees.
Go ahead, downvote. As if working in the office would have magically protected you these layoffs.
I have been working from home to various degrees since 1997 or so and I go in more when I need more work to do and work from home more when I am super busy with coding stuff or similar that can be done better and faster from home.
But yeah going to office to sit in meetings with of people in other offices is silly.
Timezone matters
You're not wrong in that companies are looking to hire in cheaper geographies but I think the remote aspect is just a small part of it. Another part is that SV comp has shot through the roof because RSUs. There are also arguably more and better people available in the cheaper geographies.
But it's not a zero sum game and there are still a lot of tech jobs in North America. AI hasn't reduced that total number of jobs.
Talented workers get talented pay -- a fact many company's and execs don't like.
It has very little to do with their literal position on earth in relation to a company's real estate foot print.
Workers need to be more intelligent and what was that word you used in your comment... "collective."
The U.S. managerial class has been GORGING off worker's labor for decades and workers are about to have nothing to show for it.
I've worked with absolutely fantastic developers in Argentina, Poland, Romania and Ukraine. You're right, talented workers get talented pay, but talented pay in Romania is a lot less than talented pay in San Francisco. Teams are used to working entirely remotely now, videoconferencing software is great, and teams mostly align to US timezones (not entirely, but for example Eastern European workers shifted their day a few hours later so we get at least 4-5 hours overlap every day).
I think companies are a lot smarter about outsourcing now than they were 15-20 years ago and focus more on developer quality, but I've absolutely seen that a company will be much happier to hire 2 excellent senior devs from Poland for what you could get for 1 junior college hire in a high cost of living area in the US.
> It has very little to do with their literal position on earth in relation to a company's real estate foot print.
Which is why they will happily pay for talented labor in other positions on earth.
Here’s the thing though. People have been trying to outsource software development since the late 90s. Every time I’ve been around offshoring efforts wherever they were implemented, a few years later onshoring would happen again. It turns out time zones, thick Indian accents, and poor quality control have been and still are major obstacles to overcome.
15-20 years ago SWE work was brutal and paid OKish, but not great if you calculated the hourly.
Then the era of free VC money came, culminating in the pandemic boom, where people were crash coursing JavaScript to land a remote job doing 4 hours of work a week for $170k.
The pendulum is now swinging in the other direction.
Additionally, companies can’t just hire individuals in other countries. They have to set up business entities and that costs significant money to do. It’s why they mostly work with outsourcing companies, who often do have offices where these people come in and work.
No one is talking about outsourcing to body shops like the old times a decade ago. I went from employing offshored devs and IT talent competing with low-effort body shops, to now competing directly with the largest names in the industry for those workers.
It is not hard at all to employ people “directly” in most countries. Plenty of global payroll services that will handle this for you at a small scale, and you just stand up a local “office” with a cheap office manager and attorney on retainer if you outgrow that. The legal entities and structure might be opaque, but the end result is effectively direct employee as far as anyone working with them is concerned.
Offshoring does not mean what it used to. These people are treated and expectations are the same as anyone employed from Iowa or California. The largest difference and source of friction is timezone.
Reliable internet is more or less ubiquitous in the entire world now in any even moderately sized city. It’s not 2005 any more. This is utterly a non-issue. And this was before Starlink. A $200/mo paycheck bonus usually covers this anywhere I’ve done business.
I’ve worked with hundreds of folks around the world at this point. The HN take on outsourcing is so ridiculous to me, and explains the hubris you see here regarding remote work.
High pay has been a mixed blessing for the tech field. For every aspiring top mathematician or physicist who’s been tempted by the pay and relevant new problems, I often feel like we’ve gotten 10x as many people who would otherwise have been uninspired doctors and lawyers or top business majors.
It wasn't some digital nomad work in my jammies lifestyle thing. It gets tiresome to hear: "see it's all your fault"
One of my friends from school has long worked in robotics. He showed me a machine he programmed to cut glass or something at a Volvo factory. I asked him about outsourcing back then. He told me his job required being physical plant to assess, program, and monitor it. He was going to stick with that over the higher-paying alternatives that he believed would be outsourced or automated more easily. Today, outsourcing only threatens work opportunities for one of us.
The era of American developer exceptionalism is over.
Talent abroad has access to the same tools, education, and increasingly, network. American engineers will be replaced wholesale with overseas engineers that cost a fraction of American labor.
You can hire talented React engineers for $50k that will work harder than their American counterparts.
It's not just React. Overseas markets have DevOps, SREs, embedded, systems engineers, you name it.
For years Americans joked that overseas labor was subpar. It's not, or at least it isn't in today's world.
Everything I look into as far as 2nd career goes has a very high barrier to entry and then there's the ever-present ageism barrier to fight, too.
I'm fortunate in that I have enough money to maybe just be retired, but most of you aren't anywhere close to being that financially independent. It's going to be ugly.
I knew for years that the offshoring blitz would finally reach critical mass and I was correct. Now it is an economic conflagration.
So yeah it’s not surprising that jobs are going offshore.
Jobs that have a high 0-1 component will still be in the US but jobs that are more 1-n may be offshored.
And developer costs had gotten so crazy, this was the perfect storm.
It's not like they don't have money, it was an insurance company with $1bn in quarterly profit. Market is extremely unfavorable for IT talents currently.
I guess folks really did think that location had nothing to do with Silicon Valley paying 4x or more the worldwide average. That there was simply no talent anywhere in the world who could even compete at their level. No project managers anywhere who could do the job for less than $200k/yr.
One the dam broke and it was clear that remote work could be productive it simply opened up a rather insular industry to extreme worldwide competition. Far more smart and talented folks out there in the world than many anticipated.
The hubris was (and is) crazy to me.
The industry suffers from fads and cargo-culting. Stealth diffuse offshoring is currently in, it's going to take a while for the downsides to percolate: I suspect will mostly boil down to jurisdiction impedance mismatches. What does it mean for employers if it is 1000x harder to extradite an employee to Delaware (or whatever jurisdiction is agreed to in their employment contract) if they clone their product from code? Without e-Verify, what are the chances that your remote employee is a North Korean unit, or working for your competitor? Boom times ahead for corporate intelligence - as well as the traditional kind. How many security teams and policies are geared to face Iran as an adversary? An unstated e-Verify benefit is free FBI COINTEL
That's a pain management are willing to inflict upon /you/
CTOs are aware of all the tradeoffs
Doesn’t it affect big tech companies? Only startups?
I would guess the opposite, with big companies being much more savvy and influenced by fiscal incentives.
One thing also contradicting the "AI can do it" argument, is that a business's playbook is to rather expand the work force in order to multiply the effect of technology. Yet they lay off in waves.
There is no dissonance, just a disguise: Several big tech companies in the US cut their workforce, in the US, while expanding it in countries where talent is cheaper.
Paradoxally, junior folks have it even worse. It has become very difficult to land a job without experience, again, in the US. It all makes sense, if you replace US based senior staff with Junior staff on different time zone, and having a different culture, you are left with nobody to mentor and supervise junior so staff.
I still can't explain how ending up with more Junior staff, offshore, less senior staff and little to no Junior staff locally will pay off in the long term.
I guess I will figure that out, but for now that's one piece of the puzzle I can only call an "economic downturn" outlook.
Salaries for tech are 2x what they should be just due to housing costs.
Seattle used to be where bay area companies relocated teams to because the cost of living here and salary expectations where 2/3rds the bay area, but that ship long since sailed.
Asinine short sighted city councils will be the death of all our jobs and also the death of the only remaining industry that America is competitive in on a large scale.
High salaries that exist just to pay for high house prices hurt everyone.
Public press paints picture of the firm laying off due to adoption of AI. Yet, I can assure you without doubt its not because of it but due to continued offshore.
I think we are at the brink of opening another large campus in India, not sure which city. Oh another thing is these full time position in NA were eliminated and replaced with employees from TCS.
My own boss is against it but its coming from higher ups. I have no idea what's driving them. We made billion in net revenues last year and we are actually part of the org that is in 'cash cow' category - upper management literally said this too in our town halls
Consider companies like Amazon, Tesla, who (initially, for many years) prioritised long term growth over short term profits. Look at how long they lost money for. You do need a CEO who actually has some vision.
https://www.futuriom.com/articles/news/why-we-dont-believe-m...
https://www.cnbc.com/2025/11/04/white-collar-layoffs-ai-cost...
Oh I totally agree with you, this was just the first thing that came up in a Google search. The criticism of MIT’s study has been going around widely, this site isn’t the only one saying that this study lacks objective metrics.
Students getting lazy, or dropping out of subjects entirely because they don't think they have a future in them.
Depression and a general feeling of despair. I see this in programming communities quite a bit - people who see LLMs as an existential threat to their careers and that they have wasted their lives getting good at something which is now being devalued.
"ChatGPT psychosis" - where people talk to LLMs and have unhealthy thought patterns reinforced by them to disastrous ends - gets a ton of coverage. But what about these milder but still meaningful effects where the very existence of AI disrupts people's future plans and self-worth even if they're not using it at all?
When we’re all brain rotted and unemployed, how will we spend money on corporations? Its a spiral.
https://podcasts.apple.com/us/podcast/conversations-with-tyl...
It was a bit towards the end, I think.
I use mobile services timeboxed and in conjunction with blockers for certain services. I also went back to use old-school pencils and paper for work whenever possible. It is helpful - and fun.
Blocking mobile internet on smartphones improves sustained attention, mental health, and subjective well-being: https://academic.oup.com/pnasnexus/article/4/2/pgaf017/80160...
Brain Drain: The Mere Presence of One’s Own Smartphone Reduces Available Cognitive Capacity: https://www.journals.uchicago.edu/doi/full/10.1086/691462
If others are slacking, it's an opportunity to level up and stand out. Also, IMO there are market forces currently reshaping the jobs landscape, it's not only AI, I don't even think AI is the main driver.
And this isn't about AI (well, not primarily anyway). It's offshoring, offshoring, offshoring.
IMO, what's taking place now is absolutely transformative and the world economy is in the process of being reshaped. It's not just tech jobs that are being offshored - we're just one of the first/early movers. Many other professional/white-collar jobs (accounting, etc.) are also getting offshored at an accelerating rate. And it's happening all over the western world - it's happening in the US, it's happening in Australia, Canada, the UK, etc.
And unlike previous periods of mass offshoring, I don't think the jobs are ever coming back.
These new tech companies/existing companies were not here for the first wave of offshoring engineers many years ago. basically, the product/service degraded and they brought the product/service back onshore.
It's a cycle that will repeat. Product degrades, there will be public outrage, then they will onshore the product to fix the problems caused from offshoring.
For one, many of these companies are now used to their tech teams being remote. The tools, culture, infra, etc. over the last ~5 years has all become remote which lessens the shock of going fully offshore.
Two, many tech teams in the western world are already partially offshored and have been for some time now. I know where I worked, a reasonable % of the team was already offshore in low COL countries (India, etc.). What's happening now is just the expansion of that cost saving after initial testing of the waters was successful.
Three, the quality gap between offshore teams and their western counterparts is now much smaller, and AI will be used to lessen the gap even further (along with just throwing more bodies at each problem which you can do when your salaries are 1/3rd of what they are here).
Four, many products/services now have captured markets with strong network effects, which means they can weather a heavy degradation of services with little to no loss of customers. It's called enshittification, and businesses are doing it now because they absolutely know they can, and get away with it.
I guess you’d expect some staff turnover, but not mass layoffs.
Opening in a private window solved the issue, however I'm pretty sure I don't regularly read anything on this site (maybe never was an overstatement?).
Over the last 20 years of tech, the giants have taken the smartest folks out there and put golden handcuffs on them. You could hire up all the smart folks and put them to work, or leave them out there and have them compete with you. With the launch of cloud providers and (expensive) dynamic scaling the problem only got worse. Think about hiring in the pandemic. Every one at home, with a stimulus check and nothing to do. Rather than a flurry of new software you got mass hiring.
But now we are in a capital intensive hardware cycle. Where in order to compete you need to have lots of $$$$ as well as software know how. It does not matter that there are smart people out there, without hefty backing they wont get very far.
I suspect that software is about to enter its "punk" era. We have software for small businesses that will help with accounting, HR, customer service, and cloud providers are starting to see some interesting competition. Much like the old punk poster showing you 3 chords and telling you to start a band it is entirely possible to find three friends and start a business that makes 1-10 mill a year with little effort and lower costs. The moment you stop thinking "unicorn" and start thinking "sustainable" the economics shift radically.
340 millon Americans. Reach 10k of them at 10 bucks a month and you have a business that pays for a couple people to live comfortably.
I don't think that ChatGPT coding is valuable but rather it's ability to tutor people and guide them towards idiomatic patterns.
Obviously layoffs correlate with AI age, but it's most definitely not AI replacing jobs, not yet. Even in 2025 stories about a job fully taken by AI need to be scraped for, and it's almost always about non-SWE jobs. And in 2023, when the first layoffs already started, models sucked and none of the existing tools and agents even existed yet! But if you search for example for headcount growth in India/Mexico, the numbers can only be described as "booming".
I don't know what exactly is going on, but it's pretty obvious the companies are moving offshore or simply doing less work, and for some reason need to lie about why.
Not a single mention of jobs being moved to India including tens of billions of investments in new offices.
Backing links from a quick search earlier this week when an obviously Indian HNer tried to deny this was really happening:
Microsoft announces US $3bn investment over two years in India https://news.microsoft.com/en-in/microsoft-announces-us-3bn-... (Jan 2025)
Google announces $15B investment in AI hub in India https://apnews.com/article/google-artificial-intelligence-vi... (3 weeks ago)
[Indian] ex-Accenture CTO named Google Cloud’s Chief Product https://www.hindustantimes.com/trending/us/who-is-karthik-na... (last week) (a lot of people speculate they named an Indian Accenture guy to move as much as possible to India)
Big Tech giants defy US-India trade tensions, record strongest 12-month headcount growth in India in 3 years https://www.moneycontrol.com/technology/big-tech-giants-defy... (September)
https://www.reuters.com/world/india/openai-launch-first-indi...
https://www.anthropic.com/news/expanding-global-operations-t...
and so on. Something very crazy is going on.
NOTE: I am not American and this doesn't affect me directly.
It has so become huge there, that is getting to similar size as the traditional call centre industry, hundreds of thousand of people work on curating data for model training.
Google hiring a senior lead who can scale low skill business processing massively is just inline with that.
Any compute investment is largely local . Internet and Electricity is expensive and erratic, there is red tape, loads of import duties, there are dozens of cheaper places for cheap data centers.
All this to say it is not the kind of outsourcing you think it is.
—-
There is a some amount of high tech outsourcing happening, those are driven by challenges in getting even a business visa in last 5 years.
those numbers are quite small in aggregate won’t be news worthy.
India simply does not have that kind of high quality talent pool .
The education system is both expensive and very poor. There is shortage of qualified teachers, even bigger shortage of good ones. Students and parents just want to get a “degree” /maximize scores to get a job, There is little interest to learn.
Given that this is presumably all data used for RLHF'ing, I wonder how much of this is responsible for things like LLM "sycophancy" or the issue of "hallucinations". What if the reward-hacking entity isn't the LLM/optimizer but the human annotator in the loop.
Then why are they giving one year of free accounts for developers to the whole country? Remember they have about one million CS graduates every year. Do you see the scale of costs?
Same goes for the OpenAI and Anthropic office announcements posted by the GP, these are sales offices with very little tech employment or outsourcing happening.
Re “investments”: these are also not what you think. I remember when Amazon was partnering up with schools in the US to provide “machine learning education” where said education was thinly veiled ads for managed ML services like Sagemaker and Rekognition. I wouldn’t be surprised if the $3B investment from Microsoft was primarily free courses and certifications to cultivate a market, as well some annotation work (I know that Amazon has been hiring quite a few people to do this kind of work).
The thesis here is that current gen tools should enable 2 people do the work of 10 and LLM models get paid the budget of 2-3 devs.
The outsourcing org comes out ahead with spending only 70% as before (salary of remaining devs will naturally increase)
——
India is by far the largest market for this. 1yr free is for same reason any VC funded company discounts their product or gives it free first, or Microsoft famously rather have you use their software pirated than not at all.
This is only to get adoption, if your workers only know how to code with Claude or Codex then you are going to buy those tools as a company
Nobody is going back once you get them to change. OpenAI et al hope to capture 30% value of the IT outsourcing sector while making to cheaper by 30%. There is no free lunch.
——
The number of CS “grads” are not a useful metric for India.
The industry shifted to hiring other engineering streams to any grads with aptitude to learn a long while back.
The vast majority of them don’t have any programming skills, don’t have critical thinking skills, are poor at communication.
They are unemployable. Don’t take it from me, NASSCOM says this too.
The degrees are largely from mills that charge up to $80k-$100k (med degrees go even higher) promising jobs but don’t teach much.
Seems to be more and more common these days in general :-/
Having dealt with A LOT of Accenture “devs”, how do I leverage the barn and then some to short Google!
Not that I believe Trump actually cares about US jobs, let alone thinks strategically about the importance of a strong software sector, but you might expect him to at least do another mafia-style shakedown of the companies outsourcing jobs. Heck, you might think he'd even care about all the lost tax revenue, by having developer salaries going to India rather than staying here.
They might use it for the next elections, all the white collar jobs are going out of US, they will bring it back if you elect them again.
I'm sure Google, Meta and hundreds of other companies will pay big $$ to make sure nothing comes out of it.
I just don't understand this need to rush, software already moves fast without AI tools. Software developers already make a lot of money for their companies that their salaries are easily covered. Realistically people could develop software 10x slower than now and the world would still keep progressing.
1. How will tangible assets generate profit net of near term capex requirements and interest on debt?
2. Why wouldn't payroll shrink as a result of the increased AI capabilities emerging from the capex spend?
3. If AI lives up to the hype & given recent news that public backstops are being requested, why shouldn't the US quasi-nationalize cash strapped players and distribute equity to every American?
4. As NVDA and AAPL local models and local compute eat into utility and base automation business, how do edge players maintain profitability without pricing capabilities well beyond the affordability of SMBs and individuals?
> Those expenditures may be approaching $1 trillion for 2025, while AI revenue—which would be used to pay for the use of AI infrastructure to run the software—will not exceed $30 billion this year
While it's clear that the author is summing up the spending from the big players, it's not clear to me that their math is right for revenue. Yes, OpenAI, Anthropic, Thinking Machines, SSI, etc. have pretty limited direct revenue (including zero!).
But this comparison assumes no revenue growth for other top computing users. Some companies are certainly saving money on some tasks and increasing revenue, particularly in fields like customer support. See the confusing figure in section 5 of https://hai.stanford.edu/ai-index/2025-ai-index-report/econo... .
That chart is by number of respondents and not weighted by revenue. Like the MIT study, it would not be surprising that "just pipe this to an LLM" isn't enough for most fields or companies. But a few have likely made material improvements.
10% of respondents saying they've seen a >10% revenue gain could be substantial, if they're bigger firms with high leverage in computing.
Edit to add: the comparison also makes a classic "GDP vs market cap" style mistake. Capital expenditure has multiple years of useful life. Revenue is annual. You'd want to compare depreciation vs revenue.
So either their business is a dead end, the inefficiency is at the management layer, or AI isn’t actually making workers more efficient.
Its why I'm not the biggest believer in "Jevon's Paradox" when it comes to software. Most software projects scale and as such they aren't really cost constrained. Or another way of stating this is "if the idea is good and can make lots of money the cost of dev's isn't a limiting factor" - hence why AI doesn't necessarily increase demand all that much. This is unlike say physical industries where cost can absolutely matter especially if the good is constrained already by affordability.
I think most SWE's know this - they see it with outsourcing as well where cheaper costs don't necessarily mean more software is done; but there is some "hope" that things will be OK.
I'd bet on all three being generally true.
Instead we're still talking about cost cutting. Seems the market is not focused on investing outside the AI moonshot gamble.
The other factor is, will revenues really stay the same for companies once we have AGI / super intelligence? It seems value will not accrue to the companies that are in chill mode.
What most people aspiring for a white collar career probably want is a job in IT - managing technology for business is much more stable than developing novel tech.
If the thought of needing to move or lose a job strikes you as horrible, this is not a likely career path for you.
And investment and experiments by definition include the risk of failing. In almost everything lies a survivorship bias and no one talks about the 100+ car makers that went into goldrush mode 100+ years ago. This is life. Netflix vs Blockbuster - already forgotten?
Also the "fail rate" - so what part is failing and why? What's with the 5%? If we have a look at exponential functions this might be a really good deal, if the 5% can account for the losses. After all, benefits compound over time.
I witnessed first hand in FAANG some quota hires and I believe that now that no one gets paid for contrived and artificial business advantages, we are back to a more merits based evaluation of workers.
But AI should not be written off as fancy something with no impact. That's the wrong take. Whether it will be a springboard to new jobs that compensate for losses or replacements - I am not yet sure, but tent to be in the former group. ML engineers take care of ML - something new that takes care of something new.
We will see.
AMZN for example overhired in various functions because it expected demand that never materialised. Admitting that is bad for the share price, but writing some woo about "AI and agility" will convince at least some investors to keep the faith.
Likewise many tasks I used to do with Java, .NET, Node.js, have been replaced by low code/no code tools with agentic orchestration.
Thinking that AI isn't replacing jobs is wishful thinking.
I'd be interested in hearing more about the particulars of this.
Workfront Fusion, Workato, Boomi, Power Apps, OutSystems
Basically the evolution of BPEL, doing orchestration with any kind of microservices exposed via APIs, which can themselves be exposed via API endpoints.
Now with MCP, tools and agents, some of those diagrams can be further automated.
Given all those articles with AI generated images I bet that some artists lost their jobs
Those jobs get killed.
On a side note this is why I find proponents that state people with agency will thrive in the AI world puzzling -> isn't the whole point of "agentic AI" to have "agency"?
With no advantage left (e.g. strength, intelligence, agency, etc) even if new industries come about why not use the AI for those too? Unlike previous industries where new domains needed more "brains" to drive/direct it, we have AI now. AI isn't a tool; it can for example deploy and can make decisions for itself. That's what the obsession with "agentic" is all about - replacing agency which at the moment was the very general domain that you still needed humans for.
This strongly favors the economic means of production remaining that are still scarce (capitalism rewards the scarce, not the efficient). Land, capital, social connections/nepotism, etc. Logically people without these will be less economically and socially valued in general - I hope I'm wrong. The current productive class have the most to lose from AI.
AI IMO breaks meritocracy and skilled based work long term assuming they succeed. Even if not in the next decade, and not the current crop of companies pushing it I'm sure AI will eventually cause this outcome.
If you've been outsourced, ask your representative to support the HIRE Act: - Creates a 25% tax on outsourcing payments - Creates a “Domestic Workforce Fund” for apprenticeships/workforce development. - Prohibits companies from deducting outsourcing payments.
Because in any country with poor worker protections, the outcome is layoffs regardless.
AI succeeds? Layoffs of unneeded roles.
AI fails? Layoffs to cut expenditure to make up for the written-off expenditure.
In the UK, if AI makes a well-established employee redundant, the employee is entitled to redundancy pay. And if the company fucks up and overspends on chasing a ridiculous Macguffin they can't just fire people without making them formally redundant.
The damage that is going to be done in the USA if the AI bubble bursts is going to be generational.
They are however pretty well trained in getting people to quit by making their job/life hell, avoiding all redundancy processes and cost
But I guess if they want another tool to help make the workplace depressing, hellish, absurd and kafkaesque, congratulations everybody Microsoft just built it into Windows and Office365 and everything else.
This year alone something like 400B was spent on investing in chips, datacenters, electricity buildouts. That's 400B that could have otherwise been invested in people.
While i don't doubt that people will find a few solid business cases for LLMs, i am on team-bubble. I don't think this investment will add 400B worth of value and I very much doubt that this 400B is any good for future growth or long-term aspirations of AGI. Investing 400B into people and (tech) manufacturing would be a solid long-term bet with benefits.
I'm not saying that LLMs and the current AIs aren't useful, but they aren't worth even close to $400B. Some of the progress, especially in the medical field, is amazing, but the spending is completely out of proportion with the gains. It only works because right now people like Sam Altman has sold people with money on the idea and now they have to continue to peddle their sneak oil, hoping that the money won't dry up.
That being said, I think we as a society are quickly reaching the point where there just doesn't exist enough jobs to keep everyone gainfully employed. There may be new jobs opening up but not ones that people in the middle or ends of their careers can quickly pick up on. I don't think this is a bad thing necessarily, just that when combined with the shitty safety net that the US has, it's a recipe for disaster. If even folks who are working full-time still need assistance for groceries, then we've already failed, and that's not even AIs fault.
I left IT for the time being, but I ask copilot.microsoft.com a question once in a while to see if we've got to AGI yet.
Today's question (I admit not so creative) was what's the market cap of TSLA.
Its answer was "Tesla Inc. (TSLA) currently has a market capitalization of approximately $429.52 billion USD. This figure is based on the latest trading price of $429.52 per share"
It allegedly used a web search to find this out, and it included a screenshot showing the market cap as $1.4B.
On the other hand, chatgpt.com prudently tells me to go look it up myself, because it doesn't have live data.
I guess OpenAI is winning?
But for code generation, Claude is awesome. ChatGPT is OK too but I've had much better results with Claude. Weird things happen if a new version it a library came out with breaking changes though. You'll have to write rules or ensure that the new apis are in context.
But that's kinda my point. It's not good enough to replace a human, but it is good enough to make a human more efficient.
The spending-revenue gap is real. Hyperscalers are projected to spend $300-550B on AI infrastructure in 2025[1] while generative AI revenue won't exceed $30-40B [2]. Amazon's capex jumped from $48B in 2023 to $84B in 2024 to a projected $100B+ in 2025[3], that's capital intensity doubling from historical norms of 11-16% to over 22% [4].
But here's what the article misses: this isn't financial desperation. When Amazon's CEO announces 14,000 layoffs and explicitly states that AI will enable "fewer people doing some jobs"[5], he's revealing the strategic logic — show me the incentives and I'll show you the outcome. Companies aren't cutting jobs despite AI spending; they're cutting jobs because they know AI spending will pay off.
To be clear, the article treats the spending-revenue gap as evidence of irrationality. But infrastructure buildouts always precede revenue: railroads looked insane before they transformed commerce, electricity grids consumed massive capital before delivering returns, the internet required enormous infrastructure investment before creating trillion-dollar companies.
What's different now is companies are pulling the future forward. If we take this article at face value which I can appreciate is a BIG “if” then AI is already automating 25% of tasks and delivering 10-55% productivity gains[6] so they're not waiting for AI to replace jobs organically. They're cutting headcount now to fund the infrastructure that will make those cuts permanent.
More broadly, this is rational capital reallocation in a winner-take-all race. Companies that don't build AI infrastructure won't gradually decline, they'll lose competitive positioning entirely. That's why Meta is using off-balance-sheet financing for a $27B data center[7], why Oracle is borrowing $25B annually despite already carrying 450% debt-to-equity [8]. They're all-in because the alternative is obsolescence.
The real story isn't "spending causes cuts" it's that AI infrastructure commoditizes human expertise, the complement to compute infrastructure. Companies are trading labor costs for compute infrastructure because they've correctly identified compute as the new moat. The job cuts aren't the price of spending on AI; they're the business model shift that AI enables.
The article is right that we're not seeing mass AI job replacement yet. But the job cuts are happening in anticipation of replacement, not as an unfortunate side effect of spending. That's not desperation just business strategy.
-- 1.(Morgan Stanley: https://www.datacenterdynamics.com/en/news/morgan-stanley-hy...) 2. (Grand View Research: https://www.grandviewresearch.com/industry-analysis/generati...) 3. (CNBC: https://www.cnbc.com/2025/02/06/amazon-expects-to-spend-100-...) 4. (Cerno Capital: https://cernocapital.com/accounting-for-ai-financial-account...) 5. (CNBC: https://www.cnbc.com/2025/10/28/amazon-layoffs-corporate-wor...) 6. (PwC: https://www.pwc.com/gx/en/issues/artificial-intelligence/ai-...) 7. (Fortune: https://fortune.com/2025/10/31/metas-27-billion-bet-turns-ai...) 8. (The Register: https://www.theregister.com/2025/09/29/oracle_ai_debt/)
They think they know.
Even if generative AI lives up to its hype, with current US administration there's no way America is going to lead the race for long. There's just not enough energy available, when those in power oppose developing many of the energy projects that make most economical sense.
right now I see even old GPUs like V100 are still popular. Maybe the old GPUs will shift to the countries with cheap electricity?
Assuming that GPUs power efficiency will increase, the same will be true about them.
I literally just said that on Reddit:
"They [big companies] were already laying off lots of people and hoarding around 100 billion in cash. This has nothing to do with AI. The elites have been at that for some time for who knows what reason. If anything, they didn't start spending money until the AI boom.
You could say AI is destroying jobs for a different reason. The leaders of the companies believe in gold rushes more than a steady stream of investments into forming and growing actual businesses. They didn't or don't believe America is worth investing in. They're simply about extraction."
By the time the AI bubble bursts, those that have investments in those companies would have already exited.
Then in 10 years time, they will look back at how they scammed the back off of the layoffs with AI driving the entire problem.
I wonder if that's true; I'm not in the US myself so I can't exactly just go have a drink in a place with lots of devs to try to find out.
The reasons are somewhat obvious:
* World economy in general and the US in particular is a rollercoaster, with the current administration being apparently dead set on flip flopping on every decision it makes, and always making extreme decisions. That's not a good time to invest. Hiring juniors is investing.
* AI not necessarily replacing the jobs, but that's not actually relevant: AI has already torpedoed the general notion that 'if you have investor money you gotta spend just hire a bunch of folks; # of employees is the primary yardstick to check company size / success', whether AI works or not. If the boss tells a VP to 'use more AI to get a handle on hiring practices', then they're going to stop hiring juniors because it looks like you're outright refusing a direct order if you hire a bunch. Even if AI 'employees' are useless, you are strongly incentivized not to hire juniors in such an environment. Juniors both lost the job opportunities stemming from companies just hiring folks because they have enough cash to do it and no good idea on where to spend it, and the downside of looking like you aren't on the AI hypetrain, if you hire juniors.
* There's evidently been a rather massive push in particular during the previous administration to get folks from dead end jobs into IT, so there's now an overwhelming amount of junior devs, and many of them didn't naturally get drawn to the profession; they were told it's an easy way to get a steady job.
* Even though there's some downturn/uncertainty, seniors/mediors aren't being fired because companies still remember how expensive and difficult it was to (re)hire dev teams post COVID. But that just makes the market for juniors even worse and makes it harder to hold out hope. When everybody is getting fired, then once the economy is in better shape you stand a good chance. But that's not happening; those mediors and seniors are continuing to get job experience whilst the juniors aren't.
Those 4 combined: Sure, yeah, I can imagine your average junior dev's odds to get hired are at this point well into the single digits. But is that actually true?
OTOH, if it is only dreams of AI, manifested as AI spending, or CEOs laying people off thinking that AI will soon (even if not today) be capable of backfilling them, then this may well backfire, and will be "reversed" if demand is less than forecasted and/or job-replacing AGI doesn't materialize, and all we get is productivity tools, useful mostly for a narrow band of jobs.
Incidentally, I think Karpathy might be right that there is pretty much only one job that seems it actually could be replaced by AI today, at least potentially, which is call center tech support, or customer support, staff, who are dealing with a narrow domain and just reading off a script.
Are the numbers really that bad?
Google has been claiming that their cost per AI search query has dropped by over an order of magnitude.[1] They're presumably reaping the gains of not cranking up a really smart model on dumb questions.
[1] https://arstechnica.com/ai/2025/08/google-says-it-dropped-th...