Wound sg ever consider using image recognition to aid in search functions?

A lot of the time I need to reference other assets that maybe appear in files that aren’t named primarily after that design ie (say i need to ref our shows established skull design but it was established in an asset of a book or poster and i don’t know the name of the particular asset it was in)

It would be super handy if sg utilized basic image recognition to add meta tags to assets based on what was in the preview image. Would that be useful to anyone outside animation design?

It would probably need to be a separate search function to not interfere with the regular asset search but I thought it would be super handy? I’m just a designer so forgive me if this is an impossibility but captcha already uses this so its probably pretty accurate at this point?

I can’t speak to SG’s roadmap, but what you’re describing sounds technically doable, and AFAIK there’s at least one product that has the auto-tagging-via-ML feature you described: https://das-element.com/
Something like this could also be accomplished via a custom add-on to SG (e.g. via webhooks), although that presumes your company has an in-house development team (or is willing/able to out-source development).