Ask HN: What is the burden of proof for a bug reporter before you investigate?

If a bug is reported by the general public in a system you're responsible for, at what point do you go from "not my problem" to "I should probably take a look", and where do you think the line should be drawn?

There are two extreme answers to this question, one is "I'll investigate any bug report", which is excellent customer support but puts you in danger of wasting your time. The other is "I'll investigate when you prove to me that this is a problem in my system and make it easy for me to reproduce".

I'm not sure if I'm skewed more towards the former than most engineers, but I find it super-frustrating as a consumer when I feel like I basically have to find the root of the problem before support will take me seriously. At that point I've effectively done their job for them.

(This is top of mind for me right now because I've discovered an obscure bug in either my RME audio driver or Apple coreaudio, but getting somebody to look at it is like pulling teeth. But I also had a similar experience when I reported an issue in FlutterFire a few years ago, where my "minimal reproduction scenario" was rejected because, despite being <10 lines of code, it involved a 3rd party lib. End rant)