I don’t think it’s especially likely that you’ll find consistently interesting, well-reasoned discussion through any platform bringing together anonymous strangers in an ephemeral manner.

I think consistently interesting discussion has shared stakeholding as a foundational aspect - participants need to actually care, either because the discussion is a product of some commitment they’ve each made (e.g. reading something for a book club), or because the participants are familiar with each other and the outcome tangibly matters (e.g. a physical town hall meeting).

Otherwise, I think you’re more likely to get what you’re looking for from adopting some tangential hobby and having those discussions with the friends you get through that.

Ah, of course - that’s unfortunate, but thanks for the pointer.

Not well versed in the field, but understand that large tech companies which host user-generated content match the hashes of uploaded content against a list of known bad hashes as part of their strategy to detect and tackle such content.

Could it be possible to adopt a strategy like that as a first-pass to improve detection, and reduce the compute load associated with running every file through an AI model?