News for hackers. Stuff that matters.

A comment on Why National Labs are investing (heavily) in AI. Parent comment

Re:Why National Labs are investing (heavily) in AI by amelius on Tuesday May 13, 2025 @07:37AM (#43971766)
The problem I have with AI is that you cannot reason with them after they make mistakes. They always reply with "Good point!" and then they give me another iteration of bad fixes. You really have to know when you should give up and do things manually.
[ Parent | on HN ]
Re:Why National Labs are investing (heavily) in AI by chairhairair on Tuesday May 13, 2025 @08:07AM (#43971981)
Gemini 2.5 is much more confident.

Anecdotally, I think this behavior is undesirable for most commercial LLM use cases. I have several friends that have complained about Gemini’s “back talking” and prefer ChatGPT’s relative sycophancy.

[ Parent | on HN ]
Re:Why National Labs are investing (heavily) in AI by amelius on Tuesday May 13, 2025 @08:56AM (#43972427)
I was talking about ChatGpt, but other llms also have this problem. I don't really care how polite they are, etc., I just want to get the job done, but the llm often gets stuck at some point.
[ Parent | on HN ]