Re:Researchers Tested AI Watermarks–and Broke All of Them
by jacobr1 on Friday October 06, 2023 @05:56PM
(#37796934)
I think you'd need device levels keys. You couldn't trust any particular image ... but you could perhaps know where it came from, which you gives you a better substrate upon which to infer trust.
[ Parent
| on HN ]
There are no comments on this comment yet.