‘Protected’ Images Are Easier, Not More Difficult, to Steal With AI

Date: 2025-06-09 15:35:49

New research suggests that watermarking tools meant to block AI image edits may backfire. Instead of stopping models like Stable Diffusion from making changes, some protections actually help the AI follow editing prompts more closely, making unwanted manipulations even easier.   There is a notable and robust strand in computer vision literature dedicated to protecting […]The post ‘Protected’ Images Are Easier, Not More Difficult, to Steal With AI appeared first on Unite.AI.

Sources:

More From: