Post #4431916

2026-08-07 10:05 UTC

LLM fatigue is real My day job basically requires me to use LLMs (or, it is strongly encouraged) on the software that I develop. Now I understand why LLM fatigue and LLM burn-out happens. It is because you constantly have to babysit the LLM to keep it from screwing things up. Babysitting is exhausting work. At one time in my life, as a 20-something-year-old, I was responsible for a group of 30 kids age 5 to 12. Every day I went home so drained that I could barely stay awake long enough to have dinner. I know from experience that school teachers perform some of the most labor-intensive work in the modern world. It is what they call “emotional labor.” When other people are unable (as with kids) or unwilling (as with a lazy roommate) to live or play or work with you as a cooperative peer or team member, you essentially have to spend mental energy keeping an eye on them and making sure they aren’t doing something that is disruptive. That mental energy is a real energy cost, it really does consume calories and make you feel tired. The worst part about it is that the LLM never really “learns.” Kids can learn to work and play with you. Lazy roommates can learn to live with you. But an LLM never really learns, and it keeps making the same mistakes over and over again. People who work with LLMs often end up keeping a directory of scripts they can deploy with a quick copy-paste so they can quickly re-explain something to an LLM. So the fatigue of watching over it and making sure the LLM doesn’t do something wrong is a constant emotional pressure, and it never gets better with age. #tech #AI #LLMs #LLM #VibeCoding #AIFatigue #BurnOut

Replies (3)

@x_cli@infosec.exchange 2026-08-07 11:22

@ramin_hal9001@fe.disroot.org Your employer is not worth your mental health.

Open ##4431932
@david_chisnall@infosec.exchange 2026-08-08 09:25

@ramin_hal9001@fe.disroot.org The worst part about it is that the LLM never really “learns.” I’ve had a few new contributors to some of my projects and the first few code reviews are exhausting. You need multiple rounds and they end up rewriting almost everything from the first submission. But the second one is better and needs some tweaks. And, after a few, you get to ones where you think ‘huh, I wouldn’t have thought of doing that, that’s neat’ and now you’re in a state where not only is it faster than if you’d done it yourself, it’s better because they’re much closer to the specific bit of the code that they’re touching and you’re trying to hold the whole system in your head. With an LLM, you’re stuck in that first-submission process, all of the time. And it might still be faster, because having someone else go and write the code after you’ve gone through all of the design iterations and feedback on early drafts might be quicker than writing it yourself.

Open ##4443484
@merospit@infosec.exchange 2026-08-08 09:37

@ramin_hal9001@fe.disroot.org I too am being forced to use it, with the same observations. Even getting it to create or modify scripts to do the actual work is troublesome. It will get something in its head and once it decides on it, it will just attempt the same thing three or more times in a row despite you telling it no repeatedly. I don't feel guilty at all swearing at it, so because of that, I prefer to think of it as a glassy eyed intern rather than a babysitting job, but the effects are the same.

Open ##4443850