Linus Torvalds, the creator of the Linux kernel and git, is employed by the Linux Foundation. This Foundation is a non-profit organisation dedicated to, as the name obviously implies, the promotion of Linux. The primary use of the funds it collects is to “help fund the infrastructure and fellows, including Linus Torvalds, who help develop the Linux kernel”. The list of megacorporations donating most of the Foundation’s funds is long.
The Linux Foundation has twelve platinum members, which donate $500000 per year, followed by twelve gold members, who donate $100000 per year. Below these two primary tiers lie the silver peasants, who each donate $5000-$25000 per year, based on number of employees. Looking at the list of twelve platinum members, I noticed something interesting.
Of the twelve platinum companies, six are “AI” companies or companies with massive investments in “AI”: Google, Huawei, Facebook, Microsoft, Oracle, and IBM/Red Hat. Then there’s Samsung Electronics, which is raking in stupendous amounts of money thanks to the “AI” bubble. Additionally, one of the gold members is Anthropic, another major “AI” company and makers of “Claude”, the sloppiest of slopcoding tools.
Many of these companies are unimaginably deep in the red when it comes to “AI”, with very little indication they’re ever going to be able to recover any of it. The situation is particularly bad for Oracle and IBM/Red Hat. Oracle’s debt has been downgraded to one notch above junk status because of its “AI” spending, while IBM’s shares experienced the largest crash in its 115 year history only a few days ago. By the way, in the first half of 2025, “AI-related capital expenditures contributed 1.1% to [US] GDP growth, outpacing the U.S. consumer as an engine of expansion”.
Fun fact: since most of The Netherlands is effectively a swamp, most of the country’s buildings are built on massive wooden or concrete poles (piles) hammered deep into the ground until they hit something more stable than mushy clay and wet sand. Otherwise, buildings in the country would simply sink into the ground. Every Dutch person who ever lived near a construction site has heard the rhythmic kathunk, kathunk, kathunk, all day long, as the massive piledriver machines spread their gospel. I guess something reminded me of this just now.
Anyway, a large chunk of the funding the Linux Foundation, Linus Torvald’s employer, receives is coming from increasingly desperate companies frantically trying to convince a populace deeply skeptical and often downright hostile towards “AI” to spend money on “AI” before the bubble bursts.
For some reason, I thought this was interesting.
Unpopular opinion: Linux foundation is massively overfunded.
It is swimming in money, like Mozilla, to the point that it could operate for decades without ever breaking into the red or earning another penny. And can mostly operate on interest alone.
And yet countless projects and applications on which modern distro are built are operating on fumes. Even many established kernel contributions dont get compensated from the big pot.
Linux Foundation has become a Fifa-like cash cow that sucks in all the funding but never distributes it to the actual grass roots (but does to some equally well funded projects at the top of the pyramid)
Adurbe,
This is unfortunately the fate of many useful organizations, and that is the primary catalyst to why they become completely broken under bureaucracy.
Take Wikimedia, once thriving community of volunteers, is now running sham donation campaigns, since like you mentioned, they also swim in cash, and can operate for many years even if they did not take another cent in at all.
This brings in many long term employees with inflated salaries, whose sole purpose is perpetuating this descructive bureaucracy. They of course also push volunteers out.
I’m not sure whether there is a solution though, EFF, ACLU, even UNICEF became completely ineffective compared to their apparent budgets.
I agree. I think Linux Foundation should donate more money to other critical open source projects eg OpenSSL, GnuPG, Let’s Encrypt
that might be, but calling Linus a corporate shill is bit much, isn’t it?
I am basically anti-AI for most of the day… until I stop into OSnews and read one of Thom’s screeds, then something happens to my brain and I start rooting for trillionaries. PLEASE MAKE IT STOP.
It’s not that they have too much money. The Mozilla foundation wastes cash in useless stuff that has nothing to do with its intended purpose. There are tons of useful stuff that they could be spending their money in or research projects for possible future developments instead of trash projects and high salaries for useless people.
Meanwhile the Linux Foundation could pay money to someone to do that necessary but boring work that nobody wants to do.
Linus view of AI as just a practical tool is not different as his view of many other topics. For a while.
I have been following this topic for a while. I would say that this is not different from, e.g., the thing about Rust in the Kernel. The purists of C were complaining similarly. Practically, Linus is not an anti-whatever, also not obsessed with free-software, GNU or FSF. He just focuses on technology, and people don’t want to accept that because of the historical GNU/Linux link.
That’s how I feel. I watched a few interviews where he explains his opinion, and it makes perfect sense to me.
And from my own experience using Claude on a massive C++ codebase, it absolutely helps. That doesn’t mean I take whatever it writes verbatim, which is what (ironically) both the tech bros, and the AI haters seem to imagine.
Yeah, he has always been on the open-source side of the free/open divide.
i know what you are referring to, but Linus is who he has always been: a pragmatic person.
There is nothing sponsored about his opinions on AI. When used responsibly, it helps immensely. But it still needs a human supervision and intelligence to deliver the right results.
I am skeptical of AI, but i leverage LLMs for hints and assistance when dealing with tech i am unfamiliar with. And it really helps. It does not do my job for me, but it saves me hours of research.
Less than three years ago, Linus was calling out “AI” for the overblown hype machine it is, and was heavily against it being used for “real development”. Now he’s completely flipped his opinion, and is rabidly for slop in the kernel to the point of telling anyone who doesn’t like it that they can fuck off. Gee, I wonder what could possibly have happened $$ in the past several months $$$ that could have $$$$$ changed $$$$$$$$ his $$$$$$$$$$$ mind $$$$$$$$$$$$$$$$$$$$$$$$$$$$$?
https://www.linuxfoundation.org/press/linux-foundation-announces-the-formation-of-the-agentic-ai-foundation
His take was much more nuanced, and not a complete dismissal. So the “completely flipped his opinion” is a hallucination, to stay in terms of AI 😉
2 years ago:
https://www.tomshardware.com/tech-industry/artificial-intelligence/linus-torvalds-reckons-ai-is-90-percent-marketing-and-10-percent-reality
“So my approach to AI right now is I will basically ignore it because I think the whole tech industry around AI is in a very bad position (grimaces)…”
+
“In five years, things will change, and at that point we’ll see what AI is getting used every day for real workloads.”
+
“But, I really hate the hype cycle.”
Tools can be a overblown hype at first, before the real and good usecases settle.
Your own quotes don’t agree with your assessment. I said he called it out for the overblown hype machine it is, and Linus said “I think the whole tech industry around AI is in a very bad position (grimaces)” and “I really hate the hype cycle”. Today, he is saying the opposite, that it’s a useful tool and if you don’t like it you can fork the kernel or walk away, i.e. “fuck off”. If that’s not a flipped opinion I don’t know what is. And it just so happened that his opinion took this 180 degree turn a few months after a multi-million dollar deal closed where he is the most prominent beneficiary.
I do not see in what way that is a “flipped opinion” if he acknowledged the possibilities of it 2 years ago but hates the hype that it had back then (and still has – and hating that is the right thing to do – one could compare it to “foie gras” – pushing it down our throats is just wrong). in the video, the first sentence he says is “I think AI is really interesting and it going to change the world” (sorry that I did not play it already when writing my first post, I just hate watching videos and interviews so much haha)
the thing about the “fuck off” is just blunt for he does not care if you like it, that they’ll allow AI use for the kernel.
If he had said “fuck off with ai, I never want to have to do anything with it” and then he write things like he did now, sure, thats a flipped an opinion. but his statements back then sound to me like he is open to it, just doesn’t care about the hype and waits until he sees some real use (aside from “great demonstrations (rubs forehead)”)
maybe the details are lost on me, as english is not my primary language – but it really does not read like “flipped opinion” (for me)
It’s called adapting to reality. Two years ago AI was not usable. Today it very much is.
Also, the hype has nothing to do with utility.
I share Torvald’s view here. I really dislike hype as well, but the fact that there is hype doesn’t mean LLMs are never useful tools.
The thing is banning won’t stop people from actually using AI and then submitting it as their own work. So I genuinely want to ask if people think it’s better if we know when and where AI is being used or if it’s better as a don’t ask don’t tell situation such that those using AI don’t tell anyone?
Today’s LLMs are finding software faults that had gone undetected by humans. I know some people hate AI with such a passion that they want LLMs uselessness to be a foregone conclusion. However if a human did the same work finding/correcting faults it would be considered good work and even be cited as an example of FOSS working as it should. Banning the “good guys” from having LLM tools creates a problem because it gives “bad guys” a vulnerability information advantage. Evildoers will have no quips about using LLMs for their purposes.
I don’t blindly subscribe to a pro-AI agenda and genuinely share many concerns over abuses. I care about my fellow humans and am passionate about creating a better life for everyone. Many of my reservations around AI come down to exploitation by elites while throwing the working class to the curb. Debates with this line of reasoning strikes a real chord with me because I am very emotionally invested in the human condition.
Honestly though I find that a lot of the anti-AI coverage pushes a denialist argument, which I don’t find to be compelling at all. The cat is already out of the bag. LLMs and other AI are here to stay, even with a recession it’s not going away just like the internet didn’t go away after the dotcom bust. It’s so important for us to stand together on a grassroots level to call for AI to benefit all of humanity rather than just the elites. The seeds we plant today will grow into the forests we have tomorrow. However when we keep pushing an anti-AI agenda and refuse to plant our seeds, we’re not truly going to stopping AI, rather it’s going to be left to the elites who’ve been planting their own AI seeds. IMHO it is a big strategic blunder to give up influence over the most powerful tools we’ll have in the future.
Fault finding is a very good use case… This is not “slop coding” as the LLM is not producing any code, it’s just pointing out bugs.
Of course it still needs to be used properly, as it will undoubtedly generate false positives and could result in maintainers being deluged with bogus reports. Where it works is with individuals using the LLM to pinpoint potential bugs, and then manually verifying them before submitting the results.
If you don’t like the idea of LLM writing code then you can always develop your own manual bugfix once the issue has been proven.
bert64,
I’d say these tasks are are closely related. You seem to be suggesting projects should welcome the use of LLMs to identify bugs and then keep humans around to write and verify patches, which seems reasonable enough. It’s just that there are still people who don’t want to recognize the merits of LLMs in any context, preferring policies that would have LLM tooling be banished altogether, and that’s what I was addressing before. I think Linus is making the right call here. FOSS goals should be about approaching tooling pragmatically, adapting to tools that work (or do not work) appropriately.
It’s exactly the same on when it comes to Rust.
“Only fools never change their minds.”
Don’t know on why my answer wasn’t posted yesterday, so posting it again. It’s not about persuading anyone to change their mind, it’s about pointing out it’s the same thing. Linus supports both AI and Rust due to the same companies pushing for it. Otherwise Linus likely wouldn’t care much about either of them. But what i wanted to point out is on why the double standards? Blaming Linus that he allows mentioned companies to promote AI through Linux, but not blaming Linus for the same companies promoting Rust through Linux. Said that i agree with Linus, on where he said Linux is not a sw project. For example often some FOSS projects are not being promoted due to personal views of their authors, although said that when Linus choosing the GPL license, that was a rather big sw statement.
I don’t think that the companies pushing for anything have any impact on what Linux is thinking. Rust became a fine tool to work with in the kernel without any arms bending by a supposedly “big rust” conspiracy. Same for AI, you don’t have to be a minion of “big AI” to see the benefits of having a set of tools that can help tackle complexity in software at that level. I agree that there is a bubble and that current AI industry is not healthy, but this doesn’t remove the good points of the tool set.
Big tech is the biggest beneficiary of FOSS and as such i support they contributing financial means to FOSS development. Currently there isn’t any mandatory tax or similar involved, on where independent FOSS projects could be entitled to financial support without any terms and conditions involved, that come with big tech in current system. So we in the end have to accept certain level of enshittifaciton. For GNU/Linux this currently means Rust/AI. On the positive side, at least Linus is a smart guy, knows that some people won’t put up with that. So, being a GNU/Linux maintainer you are free to choose if you will have anything at all, with Rust or AI. You don’t even have to touch it with a stick if you don’t want to. And that i guess still makes is somewhat acceptable.
Linus is an intelligent, pragmatic guy, with more systems programming experience than the vast majority of us. His position on AI is imminently reasonable. You, thom, are an idealistic extremist, with not much programming experience, spinning conspiracy theories.
sadly you are right, Thom has become an extremist, if Linus does not say what he wants to hear then Linus is a Corpo bastard, but Linus is a pragmatist, not an extremist, he said AI was B.S. when it was, then AI evolved and now the output and usefulness is enough to change his mind, but Thom is not a programmer, so he can’t understand the evolution of the tool. AI is just a tool from the perspective of a pragmatic developer.
I second this. Thom, respectfully, your ideologic campaign against everything AI shows only one thing: that you don’t really know what you are talking about. It is clear that you have completely lost your objectivity, which is quite sad.
Appeal to authority.
I think this is just distasteful innuendo.
Linus Torvalds is being consistent with previous behavior, no need to insinuate he is being bribed to explain his current posture on AI. That he has turned around sooner than (he) expected is easily explainable: AI has become usable for software development far faster than forecasted, all nuances aside.
And it could be argued that the composition of the sponsors pool isn’t that surprising, either: it’s the usual heavyweights with some churn. Getting the Linux world to use AI for development isn’t going to save them when the bubble bursts, the way they have setup themselves. Only the ones with decent fundamentals outside AI will. The argument doesn’t hold.
(“Big Rust” as a concept is just the funniest thing)
I’ll admit I had the same thought. I wondered if his benefactors took him aside and painted a scenario where they would do to him what Steve Jobs did to Apple in the 80s if he didn’t publicly soften his stance about AI and allow AI-generated code to be used in the kernel simply for the massive PR win that would represent. The world’s most famous programmer (though Ken Thompson and other legends like him are still alive and should be more famous!) publicly approving AI coding in the world’s most famous non-proprietary software project is the perfect thing for LinkedIn techbro CEO cryptoinfluencers to make insane posts about and continue to soak up VC cash.
But it’s also possible that he genuinely softened his stance based on the much-improved quality of bug reports/fixes they were getting, as they said. It wouldn’t be the first time Linus was out-of-step with the FOSS community. He doesn’t seem to care about the systemd controversy for example. He was never as angry at Microsoft as most Linux users were (even in the 90s), and IIRC he once said he would even be willing to work there, in spite of the media’s attempts to this day to frame Linux as an anti-Windows thing when it was never invented to be a Windows alternative but a free Unix (to be specific, a free Minix, which had restrictive licensing at the time).
I think his benefactors are definitely a thumb on the scale, even if they haven’t said anything to him. And the sharp shift in tone is hard to explain purely from a “AI coding got sufficiently better since he last spoke about it” perspective. But then he is also famously brusque.
So I dunno man.