Can I claim prior art on abusing bad scaling algorithms adversarially? :-) (see PoC||GTFO 15) http://scaling-attacks.net/
Show this thread
-
I'm not convinced they've used any ML besides resnet tbh. Like, 'can we trust self-driving cars and facial recognition' makes no sense when any decent ML in those categories would probably be using generated data anyhow. At most this affects social media tagging.
Yeah, this is clearly largely academic and should only really affect web services that use AI stuff behind the scenes, not self-driving cars or anything like that.
