Re: [PATCH] docs: remove Assisted-by references

[Jori Koolstra]

> Op 26-10-2026 15:39 EDT schreef Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>:
> 
>  
> On Mon, 26 Oct 2026 at 09:04, <lyude@xxxxxxxxxx> wrote:
> >
> > On Wed, 2026-10-21 at 16:07 +0000, Troy Benjegerdes wrote:
> > >
> > > It *is* possible to make a deterministic LLM, and this is why I think
> > > it
> > > is critical to my personal future to have an LLM that can actually be
> > > understood by a human instead of a random number generator casino
> > > that
> > > sometimes puts "4" in your private key.
> >
> > I hate to break it to you but that is literally not how large language
> > models work.
> 
> More importantly than that - the whole "deterministic" part is a total
> red herring.
> 
> We shouldn't even reply to that idiotic argument.
> 
> Why? Nobody is arguing that we should stop take human patches because
> "they aren't deterministic".
> 

I am curious: suppose for the sake of argument that LLMs come to reach a
level of sophistication where they surpass even expert kernel engineers,
in technical analysis, code quality, speed, you name it, everything. Is
the Linux kernel project then done for in your opinion, because we should 
always prioritize technical merit, and LLMs just write better software
than people?

This is not even that hypothetical. In some areas of mathematics this
has already partially happened, to the point that for technical merit
alone, there is no need to let people do PhDs. Whatever they can do, is
automatable.

LLMs ARE tools. They are things that humans use to be more productive.
A better definition of "tool," I don't know. But are they "just tools?"
It is hard to agree on that, LLMs are the centerpiece of one of
the biggest economic expenditures in human history. They are also a
threat to the open source movement, because they let you essentially
circumvent licensing. They are a promise for improved drug discovery,
unprecedented access to information, and disinformation. They are a
source of hot debate in science, education, and here in the kernel
community. I think this distinction between "tool" and "just tool"
is useful to make for productive discussion.

Jori.