alt Hacker NewsThat's such an incredibly basic concept, surely AIs have evolved to the point where you don't need to explicitly state those requirements anywhere?
Replies
thunky • yesterday at 2:56 PM
Doesn't seem to work for humans all the time either.
Some of this negativity I think is due to unrealistic expectations of perfection.
Use the same guardrails you should be using already for human generated code and you should be fine.
They can still make mistakes.
For example, what if your code (that the LLM hasn't reviewed yet) has a dumb feature in where it dumps environment variables to log output, and the LLM runs "./server --log debug-issue-144.log" and commits that log file as part of a larger piece of work you ask it to perform.
If you don't want a bad thing to happen, adding a deterministic check that prevents the bad thing to happen is a better strategy than prompting models or hoping that they'll get "smarter" in the future.