People are complaining that an advanced fill tool that’s mostly used to remove a smudge or something is automatically marking a full image as an AI creation. As-is if someone actually wants to bypass this “check” all they have to do is strip the image’s metadata before uploading it.

Chatbots passed the Turing test ages ago, it’s not a good test.

Context onwhat the “doxxing” actually entailed and why it happened.

What was shared was a leaked chat transcript demonstrating the coordinated efforts of many working to silence support for Palestinian liberation.

The transcript clearly demonstrates collective actions taken to contact employers, funding bodies, publishers and journalists to censure anyone deemed to be a threat to zionist ideology. Some of these efforts were successful, the firing of Antoinette Lattouf being but one example.

Basically, they were running harassment campaigns out of a chat that used real names and someone leaked it.

The notice itself isn’t malware, but I’m not sure on the cause. For some reason the Lemmy web client occasionally tries to cache a metric fuckton on iOS Safari (and probably elsewhere but it happens silently), which sounds like something for the devs to look into.

Generally, Lemmy’s 100x better in stability and speed than it was a couple months ago when a bunch of new people started working on it, but before then it was the side project of a handful of people and it showed. EG - The infamous, three-year-old Hexbear instance managed to have the entire picture side of it go down for a couple days because someone uploaded an absurdly large, extremely low quality photo of a North Korean soldier on it. So, there are probably still some issues like that kicking around.

It can also be used as a form of semi-covert harassment. For example, in an early instance, every bit of content posted by or in support of transgender people would get hit by a barrage of downvotes from brigaders using otherwise inactive accounts. Hard to moderate that without just disabling downvotes entirely.