Re: GCC LLM policy adopted

"Theodore Tso" <[email protected]> Fri, 31 Jul 2026 22:34:00 -0400
Newsgroups gmane.linux.debian.devel.vote
Message-ID <[email protected]>
On Fri, Jul 31, 2026 at 12:22:00PM -0500, Aigars Mahinovs wrote:
> > Instinctually to me, there's so little to be gained from automating the
> > writing of ~15 lines of code that I'm not very worried about this.
> 
> You know the old joke about the TV repairman with a hammer?
> 
> ...
> - The 2$ is for hitting it with the hammer, 90$ is for knowing where
> exactly to hit
> 
> Same with AI-generated contributions, especially bugfix patches and
> security fixes. The most useful of them tend to be way smaller than 15
> lines of code, but can take massive effort to actually find the right place
> and the right change.

Exactly.  An AI-generated contribution which adds something such as:

	spin_unlock(&sbi->s_error_lock);

to a return path, or adds a missing

	sbi->s_ea_inode_cache = NULL;

after s_ea_inode_cache is freed might fix a very impactful bug,
including potentially a high severity security bug.  But **how** could
such a line be considered a copyright problem?  That makes about as
much sense as claimimg that someone can copyright a four chord song.
You can make any claim you want.  But you might get laughed out of
court afterwards....

If someone wants to make a luddite-like claim that it might put your
job at risk and so therefore AI is as evil as the mechanized looms
that caused artisans to lose their jobs at the dawn of the industrial
revolution, sure, we could have that discussion.  But please, let's
not insult our fellow DD's intelligence by claiming that such bug
fixes are uncertain with respect to copyright.

				- Ted