Oracle shed about 20k roles globally in the last year
bbc.com97 points by Lyngbakr 6 hours ago
97 points by Lyngbakr 6 hours ago
A friend met with Oracle salesdroids (his company has spent a few million with them in the last few years).
They told him everything is oriented towards AI, to the point that otherwise profitable software is not being updated and development has stopped on anything new.
Oracle has a lot of niche solutions and the one my friend’s corporation uses is profitable but not on the AI path, so it is being put into maintenance mode.
That’s makes sense in its way, if they’re convinced and absolutely certain that’s the future. By analogy, I bet there were some carriage manufacturers who bet the farm on internal combustion engines when horse drawn carriages were still popular. It would seem insane at the time for them to give up their reliable business for something risky and unknown. However, in retrospect it would seem like bad business to throw resources after dead-end product lines, even if they’re valuable, because each dollar not advancing their automobile R&D was a dollar wasted on legacy stuff.
I’m not saying AI is the next internal combustion engine. If Oracle is certain it is, though, then that sounds like a more rational move.
> However, in retrospect it would seem like bad business to throw resources after dead-end product lines, even if they’re valuable, because each dollar not advancing their automobile R&D was a dollar wasted on legacy stuff.
If each dollar you throw at dead-end product lines returns $1.20, that gives you more to spend on R&D for the new hotness. Shutting down the old business doesn't maximize spending on new R&D.
Now it's different when there's a conflict over resources. . However, in retrospect it would seem like bad business to throw resources after dead-end product lines, even if they’re valuable, because each dollar not advancing their automobile R&D was a dollar wasted on legacy stuff. If we need kstrauser to keep the old business running and we also need kstrauser to have effective R&D for the new business, hard choices are ahead. For carriage makers, manufacturing facilities would be the big resource issue; maybe you need to stop production of carriages to start working on horseless carriages because they're built in the same facility and can't share and a second location would be too costly.
I don't think the facility argument applies to Oracle. And probably not the key person argument either; line of business apps and AI are pretty different and I wouldn't expect a top person in one to be a top person in the other (although they might be).
That's a valid point, and to be clear, I'm not claiming Oracle's making a wise choice. Among other business risks with the plan is that customers who've been burned by Oracle abandoning the specific product line they needed might not be quick to start using Oracle for all their AI stuff. Like I don't think anyone likes using Oracle's products, and if you're starting your company's AI plan from scratch, that might be an excellent time to consider not-Oracle as your provider/partner/contractor/whatever.
However, if your dead-end investment has a 20% ROR, and you think your AI investment will have a 1000% ROR, you'd kinda be foolish not to throw every possible resource at the new venture. I'd bet that a lot of the people maintaining Product A could be switched to the AI project. It's going to need lots of people supporting the networking, CRUD APIs for various things, sales, tech support, legal, etc. The product itself will be different, but much of the underlying support might look the same between the two. Why "waste" someone maintaining Product A's CI/CD pipeline when they could be helping the AI project move faster?
And again, I'm not arguing that they're right. It's more that if they're completely convinced that this is the future of their company, then that could be a rational, defensible decision. There's a lot of "ifs" in all this, to be sure. And in any case, don't make the mistake of anthropomorphizing Larry Ellison.
Anecdote time; startup I worked for was bought by Oracle in 2015
Oracle was all in on spinning up a less corporate bubble that felt like the startup zeitgeist; we had Slack while Oracle legacy units had in house XMPP
Management was dumping everything into data center and cloud growth and trying to catch up
I stayed for a couple years to help my team migrate and prepare for release the product Oracle bought the startup for. Not long but long enough to get a sense of the internal story and driving themes
No doubt in my mind this is Oracle trying to get ahead in this hype cycle. By 2017 upper management did nothing but lament how far behind Oracle was.
Cut n run from everything early this time to go all in on new meme is exactly the kind of choice Larry would make given constant anxiety on display about missing the cloud bubble.
You can use AI to eliminate Oracle from your org's dependencies.
In a roundabout way, AI may just eliminate Oracle from my org's dependencies
Absolutely unhinged behavior. If orcl goes nuts over chatbots and dies that will make this strange and profoundly icky era in tech all worth it.
The good bits will be sold off. Some of their products are very valuable, even if you don't like them.
That's a good thing. While Oracle's software has it's ups and downs, the worst thing about Oracle has always been Oracle.
Yeah I don't dispute they have (had?) successful products. Just saying for a company to abandon successful products in favor of a wildly costly gamble is insane. To even arrive at the point where it seems like a "tradeoff" is delusional thinking--like, "we're going to stop supporting this successful product because the future value of this other thing is greater"... what? They've contorted themselves into believing that the future value that product is irrelevant, and all their customers whose trust they're burning are irrelevant, because... chatbots?
It seems reckless and unnecessary even if the gamble were 100% a sure thing, which no gamble ever is.
Even if AI is a game changer, it doesn't mean that your business model will have a role in it.
That's rational. Oracle has bet the farm on AI, if they don't make it work they will fail.
It doesn't seem rational to me that such a big company wouldn't prefer to diversify.
Once they made the loans they can only pay back if AI works out, no, it's completely rational to put all of their effort into making AI work out.
Either that, or to declare bankruptcy earlier to get better terms.
None of this is due to AI. That’s just the cover story that stops the stock crashing. The reality is the market is contracting and it’s doing that because the extraction and monthly financial models are hurting other businesses which are also contracting because of lower customer demand.
IIRC, WARN filings in NYC require to indicate whether lay-offs were due to AI. None of the filings of the past year checked that box [1]
[1] https://www.hrgrapevine.com/us/content/article/2026-02-11-ai...
We had the cops investigate themselves and they found no evidence of wrong-doing.
Businesses are incentivized to say it’s because of AI because it makes them look better. So not saying that when actually asked by a government entity is meaningful.
How does firing people resultant of the most disliked and unpopular technology on the planet, and avoiding marking it off on an obvious entry-point into regulatory burden "make them look better"?
Because it sounds better than saying "we're firing people to suppress wages and make even more absurd profits. oh, and we invested a collective $1.4T as an industry so far in AI but can't find a path to profitability".
Anecdotally, we (humans) appear to be more accepting and less critical of companies giving AI-related layoff excuses, vs. perhaps telling a more truthful story like "we are losing money and are scrambling to reach profitability" or "we'd like to make the stock price go up."
Though I do agree there are a vocal group of people that are loudly outspoken about AI. I would guess that those people are a minority of the total population though, and tend to skew either techy or geographically local to areas hardest hit by data center build-out.
No, the majority of the population is hostile to AI. This has been surveyed time and time again. I definitely disagree with your premise.
That may be true and I'm absolutely unfamiliar with the data, so I'll take you at your word.
Regardless, whether or not the majority of people are hostile to AI was not my main point. My point was that regardless of the public sentiment about AI, it seems that the general public is more accepting (or perhaps forgiving) of a CEO saying "We are laying off people because of AI" than, for example, "We are laying off people to save money and preserve shareholder value." (I am paraphrasing in both of these cases; obviously a real statement would be more obfuscated in corporate-speak in either case).
That seems odd to me, _especially_ if the majority of the public is hostile to AI.
> The reality is the market is contracting and it’s doing that because the extraction and monthly financial models are hurting other businesses which are also contracting because of lower customer demand.
Can you explain what you mean?
For those curious, this is a report about firings from the last year. Not a new round of layoffs:
> Oracle shed about 21,000 roles globally in the last year as the US technology giant reshapes its business around artificial intelligence (AI), the firm's latest annual report shows.
Now is precisely not the time to cut jobs, but to invest in people's usage of AI. You have domain expertise, and you are going to burn tokens trying to replace people's jobs, but the AI revolution is not at all about people replacement, its about a change of paradigm. Watch them closely rehire thousands of new employees (who won't have the domain expertise) once they see that you actually need people to operate their company :)
Oracle overcommitted on behalf of OpenAI, taking on debt to meet their obligations. Not that healthy companies are not also doing ridiculous layoffs, but in this case, Oracle does have to do some drastic things to get out of this tailspin.
Tailspin? Last quarter was their best quarter ever in terms of topline, and they cleared over $4B in profit.
Depends on which jobs; at a big enough scale you have people who are so deep in the hierarchy that their real connection to the domain is low.
Seems possible to flatten out a company, in a way the domain knowledge is kept.
That being said, I haven't seen many layoffs that actually seem directly AI related, because the main effects are not hiring junior Devs, and not outsourcing to low cost/skill areas.
> Seems possible to flatten out a company, in a way the domain knowledge is kept.
This definitely seems possible, but I've rarely seen it work. That being said, the follow on from any layoffs (reduced morale, more switchers) make it hard to be certain here.
What I don't get is why these companies are in all the rush to replace people with AI tools. Stuff is changing so fast that capital investment right now will just need to be reworked on after a few years when things settle down a little.
And that is not even considering how much it costs to run AI right now at scale AND retraining people to use the new tools AND possible disruption in existing workflows.
If I was running a large org I would actively try to slow down AI adoption in most areas until there are clear established solutions.
It's because these are public companies. Public companies are not in the business of developing good products and selling them to customers and having a sustainable long-term vision, they are in the business of selling their stock. Right now, thanks to a very effective marketing campaign driven by people who are invested in AI companies, the stock market wants to see stories about employees being replaced with AI, so that's what these companies are selling.
I easily do more than x 5 the workload I did in 2020. Replacing an employee with domain knowledge (only thing valuable now), with someone with zero domain knowledge is corporate suicide in 2026.
However, if you are oracle sending out consultants with zero domain knowledge around to help.... I would rather just use claude.
What type of work are we talking about here?
Generating PRs that can’t be tested or deployed. My team has collected a whole pile!
I’ve just been thinking about this and would like to ask those of you with more experience in the industry: do you prepare yourselves in any way for crises? Do you have any techniques for dealing with uncertainty?
Let’s say someone is entering the IT job market today. How should they prepare for the responsibilities of adulthood, starting a family, buying (perhaps on a mortgage?) their first flat, and planning for a family? I’m not writing solely in the context of AI layoffs but broader. Over the last 10 years that I’ve been in the IT industry from EU/US perspective, I’ve perceived the market as economic cycles, the sine wave of which has narrowed significantly (more often markets demands changes and instability) and accelerated in the past years.
The only sure thing is to live significantly below your means as often as possible. Becoming accustomed to life as a peasant when you could live like a king is what makes you economically bulletproof.
I can go for a decade without another dime coming in the door. If I started using DoorDash and Amazon Prime every day, I'd last maybe 18-24 months. Just barely treading water is actually a great position to be in. It keeps you honest and any amount of work you do find on top of that feels like a windfall. An extra $500/m is a lot if you can make it happen consistently.
These things certainly are not industry specific. I grew up in construction where layoffs are expected sometimes a few times per year. The basic personal finance advice helps, i.e. emergency fund, pay off high interest debt asap. Not burning bridges, building a network of real people is huge. Help people out when you are in a position to and are able to. As far as mindset, I think my Christian faith plays a big role in situations that might cause suffering. Essentially, don't make your career or your salary your identity and/or highest power in your life.
Live within your means. Save aggressively. Invest wisely.
(Source: I'm one of the 21,000, let go via a 6am email after 15 years with the company in engineering and management roles.)
There's really nothing special about the current situation. Throughout history, jobs have been rarely anything close to guaranteed.
You save a good deal of cash for emergencies, live below your means, keep your skill set up to date and face the world.
The speed of change is what’s special this time. Careers are being eliminated faster than people can reskill or change careers.
I'm not sure that's true. Jobs have moved overseas at a rapid rate in the past. What careers have been eliminated, and over what time period?
Nope, you just get up every day and do your best. If the ghouls at the top decide to axe you, then that's their loss. Humans have existing far longer than capitalism, we can survive without a boss.
Tech CEOs suddenly love blaming AI for mass job cuts
You can’t read the headlines, they’re marketing like anything else. Companies are free to spin them however they want and journalists are happy to write them however they will scare you the most. Those are the only truths.
IMO it’s not sudden, the AI excuses have been gradual and ramping up for a couple years now.
Maybe because the models keep getting better and the tools are slowly getting good enough to replace people? Keep in mind we’re talking about oracle here, it’s not like we’re talking about humanity’s best and brightest.
> Maybe because the models keep getting better and the tools are slowly getting good enough to replace people?
I am not sure we are there yet.
In my experience even the SoTA models are faaaaar away from replacing humans, maybe making them a bit faster.
The correlation between job cuts and ai growth is real. But it's related more to ai cost than ai performance. Especially in Oracle's highly leveraged case.
This conversation always gets lost in the weeds. You don't need a SOTA model to replace an entire engineer role from the time he gets into the office until he leaves.
If you have a team of 10 and make them all a little faster, you can do the same amount of work with 9. Run this out over the entire industry and it's hundreds of thousands of roles that are redundant.
Do you have any research, empirical data, or other hard evidence that shows that a model (especially a non-SOTA model) makes engineers even "a little faster"? I'm aware of anecdotes, but nothing more.
The cleanest positive study is Cui et al., because it used randomized rollout in real companies. Developers at Microsoft, Accenture, and a Fortune 100 electronics manufacturer were randomly given access to an AI coding assistant. Pooled across 4,867 developers, the authors estimate a 26.08% increase in completed tasks among users. They also report increases in commits and builds.
While that research was published in 2026, it analyzed the 2023 period. Very much NOT sota models!
> I am not sure we are there yet.
When talking about bullshit jobs like, say, taking a bill received by paper and manually extracting data from it (company name, invoice number, bank account details) to enter into an accounting program, AI is already good enough according to the pareto principle.
Maybe. But the models weren't good enough 2+ years ago. And to cite just one example, in February 2024 Grammarly laid off over 20% of its employees, citing AI. So you'll have to forgive me if I am extremely skeptical of CEOs citing that as the reason today.
It's a very convenient excuse, especially given the apparent lack of public backlash to it, compared to, say, "the business isn't profitable and we need to save money."
I doubt we are there yet, no.
But we are talking about a corporation that is one of the most sociopathic in modern business history.
Say what you want about their business practices, their products are top notch.
A few years back I heard (via a friend who knows a CEO, so take it with appropriate pinch of salt given not a direct experience), that Gartner was asking the CEOs in its group meetings how many of them were laying off people due to AI, with the question presented in a very leading fashion that implied anyone not was behind the curve and failing their business or whatever.
I think a lot of them have drank the kool-aid. They go to conferences and then some talking head from one of the AI companies gets on stage and tells them things like they don't need developers and how AI is already good enough to revolutionize the world. They buy it hook, line, and sinker and then do stuff like this.
What's even weirder is the Founder of canonical has a #hiring badge on their LinkedIn profile. What kind of message is that?
Fire and re-hire from a desperate pool of anxious workers who will accept any offer to escape unemployment.
Well, not "any". -15% seems to be the magic number looking at my network and has been like that for 2 years now.
They're going to flip us all like burger patties eventually - we're done on this side.
It's also an opportunity, from a different perspective
For me personally it was a golden year in terms of recommendation bonuses - managed to buy a laptop using just them.
They are hiring. I've seen one role listed for more than a year, and it's because of their hiring process. Here's a random job, by way of example: https://canonical.com/careers/3433732/application Search for "how did you perform", and be baffled at their recruitment strategy.
LOL, I’d forgotten about that idiocy. Imagine being a staff engineer and having an interviewer ask about your high school math performance.
1. I was in the top 0.01% in my state. (Actually true; who gives a crap; it was a quarter century ago and literally no one other than Canonical cares.)
2. You can’t prove me wrong. I can’t prove me right, because I’m not the kind of loser who has a wall of their high school accomplishments they never moved past.
3. The fact that you’re even asking shows that this is an amateur hour leadership team who cares about utterly irrelevant things.
Anyone who has that stupid of a hiring process deserves to have unfilled roles. I’m hiring for several roles right now, and while I ask a great many things, I promise you that none of it relates to their high school math performance.
What's stopping anyone else from using AI and eat into Oracle's market? Maybe know how specific for the market but at least 20K people with know how on this business are available at this moment.
I'm inclined to believe that its no longer the case of employees being replaced by AI but simply those businesses being replaced by AI and the current businesses who were able to benefit from AI are reaping the momentum they have but eventually this will end.
> What's stopping anyone else from using AI and eat into Oracle's market?
Open source databases have already been doing this for decades. You can’t just clone its products and expect to eliminate it. Oracle is driven by a strong sales culture and ruthless business strategy.
And why AI can't do all that? Apparently it can be a doctor, engineer, musician etc. but when it comes to do the jobs of certain class of people suddenly can't do that. It doesn't add up, maybe sales isn't as impossible to learn and do?
LLMs can certainly make sales faster and more efficient by automating things like updating CRM, prospecting, finding contacts, drafting emails, etc. There's a whole niche industry building AI sales tools now and some of them work extremely well. But in the end you still need human salespeople (often with lawyers backing them up) to close major enterprise deals.
A bit premature I think. Right now it is employees making room for AI investments
It's a move I would only wish on my worst enemy. Good riddance to them
I suspect these AI-displaced workers in the long run will suffer the same fate as workers affected by deindustrialization. They won’t die of starvation, obviously. But they will never be “retrained” to roles that offer them similar income, status, and leverage over employers. Instead they’ll be sucked into an ever-expanding economy of service work.
> But they will never be “retrained” to roles that offer them similar income, status, and leverage over employers. Instead they’ll be sucked into an ever-expanding economy of service work.
Which will probably depress wages of service workers - unless somehow there is a dramatic expansion in service demand, which apart from particular industries (i.e elder care), seems unlikely.
>They won’t die of starvation, obviously.
Narrator: This was not, in fact, obvious.
Out of 141000, about 15%
They will be hiring frantically next year.
These companies seem to be led by mind bogglingly short sighted people. Maybe they should be replaced by AI. It can't possibly get any worse than now, right?
Next year?
There's a whole bunch of open roles at Oracle right now. https://careers.oracle.com/en/sites/jobsearch/jobs?keyword=E... 700 of them within engineering.
An adjacent comment: "For those curious, this is a report about firings from the last year. Not a new round of layoffs."
https://youtube.com/watch?v=aooiDA-AsNo
Ah yes, fire to rehire via the model training gig economy
Have they layoff their lawyers or engineers?
where its sits with the AI bubble I think Oracle is the 1 of the major companies apart from the OpenAI that would go belly up once the AI bubble crashes.
Larry has embedded the company way too deeply into the intelligence and military apparatus to ever go belly up. He’d be first in line for a government handout.
> He’d be first in line for a government handout.
And he'd take a good chunk of that money for himself, and a good amount of that to also "donate" to a foreign, nuclear-armed, hostile country's "defense" force. But don't you dare question him.
Oracle has a huge, entrenched enterprise business that will keep them alive almost indefinitely, just as Microsoft does.
I think you underestimate just how entrenched Oracle’s database offering is in the enterprise.
I did a project for a big telecom in Brazil where they kept everything in one massive Oracle database that ran out of 2 oracle refrigerators[1]. They kept freaking logs on the database, like normal application level logs.
DBA would go around screaming if you were logging too much stuff. I assumed they had some sort of periodic cleanup because I couldn't see that stuff being practical long term.
Technically it was 2 databases, one was a read replica for longer running data analysis of course.
That telecom has at least 100 million paying customers (maybe much more). I don't see they ever moving away from Oracle, they are more likely to go bankrupt than ever leaving.
[1]: Oracle RAC: https://en.wikipedia.org/wiki/Oracle_RAC
True, but they also took huge debts to build AI DCs and not sure if the DB part of the company can cushion such a fall. According to [1] their IaaS line of business brings 4.8B USD/quarter (so say 20B/year), but they have ~120B of debt (outstanding + new debt they are trying to find people to pay for).
They are justifying that on commitments (500+B USD), but 300B of those are tied to OpenAI. So, if OpenAI goes belly up or at least doesn't follow their crazy growth projections, they would have to find the same amount of consumption quickly to repay the interest on said debt and eventually the principal.
It is a lot of money for a company the size of Oracle (~500B market cap).
[1] https://finance.yahoo.com/markets/stocks/articles/oracle-500...
That is the one bit of Oracle that can't really be further subdivided. The rest of Oracle could be diced and sliced.
Why? They’re way too big to fail. Tons of corporations rely too much on them
They are not. Maybe they will be treated as such if the bubble bursts while while Trump is in power, because of loyalties.
But Oracle could relatively easily be broken up and sold off. Essentially all of the global consulting business units could sell to competitors in the various niches, the hardware business and the cloud hosting business could be separated, etc.
Plenty of companies worldwide would be up for buying pieces of Oracle at a bargain price.
"Too big to fail" only really applies in extreme circumstances (which might happen, admittedly) and with essentially monolithic businesses (or banking).
I don't think even Microsoft is too big to fail.
And I don't think people should casually entertain the idea that really any tech industry company is too big to fail, because tech corporations that cannot die point the way to a Rollerball future.
Which AI bubble? The stock bubble one or the imagined "one day all the LLMs are going to disappear from the face of the Earth" one?
Dot-com bubble crashing didn’t mean the web disappeared.
Some of the shakier dot com companies did, though.
And the dot com bubble collapse of Nortel and Worldcom shows that even well-established companies with significant non-bubble investments are vulnerable if the companies make poor economic decisions (in both of those cases, the collapse was driven by excessively aggressive acquisitions).
Likewise, LLM's won't disappear completely with a stock market crash. But the weaker players might.
The classic autocomplete survived previous bubbles.
So the predictive autocomplete will survive those too.
There's a tiny chance that this whole mess leads to a situation where Larry Ellison dies broke. That would be amazing!
I think perhaps they'll crash more than others because they aren't AI enough. It's more of an AI financing bubble than an AI bubble.
What’s crazy is how Oracle’s free cash flows have gone from “money-printing machine” to deep in the red due to their data center investments. My guess is that some of those cuts are to try to balance that?
Oracle is still mostly generating income from an old school model, a bit like IBM and SAP used to. They have huge contracts with mega-corps bringing in millions per year for a license. The have been pivoting to other things like consulting, cloud, AI for years but I still feel most of the income comes from these original db and product licenses. As someone who worked with Oracle since 1998 I'm kind of amazed they've kept at it for so long major Healthcare, Utilities, Banks still locked in and in some cases buying new contracts.
Well you have to somehow both indicate that you are not behind in the AI race but also that you are not hemorrhaging money. That means crapping all over your org in order to fund the data center spending.
All in the all-mighty fiduciary duty to your investors of course.
I'm trying to remember which Marvel capeslop movie it was where the kid gets a bunch of Oracle professional-grade computer hardware or something as a gift. Larry makes a cameo, of course. Maybe it was Iron Man 2?
Like, mate, if you want to be a superhero in real life, stop laying off thousands of people and working the remaining people ragged. That's all you gotta do.
CA will buy the scraps.
Shed roles? They fired people, not "shed roles".
If they're not hiring to fill the roll after firing it's kind of both.