My employer pays for Claude, and my approach is to use it as a better Google search. It's often not better.
Just today it made three glaring mistakes in one session:
1. It read a file in the wrong directory, because that file had the same name as the file in the right directory. It apologized when I challenged it, promising me that it would remember to "read import statements" in the future.
2. It miscounted the number of times a function was called in my repo. It said 20, while my built-in IDE search accurately showed 17. Again, it apologized when I corrected it.
3. It referred to a variable by name that does not exist anywhere in my code. It apologized, and said it was referring to a variable used internally by one of the third-party packages installed in my repo.
So many apologies.
It's the little things like this that remind me on a regular basis just how little I can trust artificial "intelligence."
I'm amused at how many hacker news accounts saw this and immediately jumped to the conclusion that you either were using the model wrong or using the wrong model.
ive been regularly seeing this exact reaction online since November 2025, sadly :/
I still do upfront planning and then do careful review of all AI content. The behavior the parent described above hasn't happened for me since around Opus 4.6.
The most telling one is counting function invocations wrong, because that's simply not how models work anymore. They use terminal commands and Python scripts for research like that (if not an LSP, if one took the time to set up their tools most effectively).
Combined with their attitude, I have little doubt that the parent has disabled tool calls, is working in some janky Harness like chat/Duo/Juno, or is using a severely reduced or outdated model.
stack_framer · · focus · HN ↗
Just today it made three glaring mistakes in one session:
1. It read a file in the wrong directory, because that file had the same name as the file in the right directory. It apologized when I challenged it, promising me that it would remember to "read import statements" in the future.
2. It miscounted the number of times a function was called in my repo. It said 20, while my built-in IDE search accurately showed 17. Again, it apologized when I corrected it.
3. It referred to a variable by name that does not exist anywhere in my code. It apologized, and said it was referring to a variable used internally by one of the third-party packages installed in my repo.
So many apologies.
It's the little things like this that remind me on a regular basis just how little I can trust artificial "intelligence."
pydry · · focus · HN ↗
ive been regularly seeing this exact reaction online since November 2025, sadly :/
Tadpole9181 · · focus · HN ↗
The most telling one is counting function invocations wrong, because that's simply not how models work anymore. They use terminal commands and Python scripts for research like that (if not an LSP, if one took the time to set up their tools most effectively).
Combined with their attitude, I have little doubt that the parent has disabled tool calls, is working in some janky Harness like chat/Duo/Juno, or is using a severely reduced or outdated model.