Related question, does any assistant do continuous code review? I feel that AI could really help with uncovering subtle bugs, but it must be good enough not to

Related question, does any assistant do continuous code review? I feel that AI could really help with uncovering subtle bugs, but it must be good enough not to bother often with false positives and unrelated issues. I often submit my code for review to ChatGPT but 99% of issues that it finds make very little sense and 1% of issues usually just wrong, so I'm kind of disappointed, but may be I miss some good prompt or something.