News for hackers. Stuff that matters.

A comment on A ChatGPT mistake cost us $10k. Parent comment

Re:A ChatGPT mistake cost us $10k by arjvik on Monday June 10, 2024 @01:04AM (#40630337)
To be fair, if I wasn't looking for this bug I never would have spotted it. That being said, you're entirely right that any monitoring or even the most basic manual testing should have instantly caught this.
[ Parent | on HN ]
Re:A ChatGPT mistake cost us $10k by Akronymus on Monday June 10, 2024 @03:44AM (#40631173)
Those kinda issues are why I ALWAYS make an integration test with calling the same insert multiple times.

I did step into that particular trap more than once (passing the result, rather than the function)

[ Parent | on HN ]
Re:A ChatGPT mistake cost us $10k by astromaniak on Monday June 10, 2024 @01:26AM (#40630458)
Easy enough, ChatGPT can write verification code along the main line. Just ask.
[ Parent | on HN ]
Re:A ChatGPT mistake cost us $10k by rvnx on Monday June 10, 2024 @02:36AM (#40630835)
+ humans could have done this mistake as well
[ Parent | on HN ]
Re:A ChatGPT mistake cost us $10k by LigmaBaulls on Monday June 10, 2024 @02:43PM (#40636737)
I spotted it right away. If you plan to use a library in your project RTFM
[ Parent | on HN ]