If a real-life fatal basilisk image (https://en.wikipedia.org/wiki/BLIT_(short_story) ) is possible, it will be created by something like this. The method is akin to how it is developed in the prequel story: An AI that is adversarially trained and accidentally produces a visual effect that crashes human brains.

There is the McCollough effect, which while not fatal, does have a long effect on the brain.

https://en.wikipedia.org/wiki/McCollough_effect

https://youtu.be/Wm8ZoVQ_OJo

Thanks for pointing that out. Another is the https://en.wikipedia.org/wiki/Tetris_effect .

> An AI that is adversarially trained and accidentally produces a visual effect that crashes human brains.

I very much doubt that would be possible in general. OTOH why using the cumbersome method of training an AI for that? Stroboscopes and flashing lights work perfectly well for a portion of the general population already ;)

We are likely already vaccinating ourselves against them though, as we all experience the development of these weapons at each iteration. It might be a watching a kettle boil situation.

The poor bastards who who arent exposed to the internet everyday are likely going to need some protection though.

To be fair to TWDNE, just generating random faces is among the least interesting use-cases, and one could do better than TWDNE with some more work.

For example, here is a demonstration of how the new model averaging/transfer tricks let you convert a face from human to anime to fox to pony to furry in any direction by any percentage (summing to 100%): https://twitter.com/arfafax/status/1296084902928986113 (And since you can encode any pre-existing photograph into the human face model, you can do things like "what if Barack Obama but only 60% human and 20% fox and 10% anime and 10% My Little Pony?" Impossible-sounding, yet, there you have it.)

The pony/furry models are the same ones used for http://thisfursonadoesnotexist.com/ and https://thisponydoesnotexist.net/ , incidentally. (TFDNE/TPDNE came out much nicer than TWDNE because of better cropping/aligning, longer training runs on entire TPU pods rather than a few GPUs, and transfer learning.)

We're also still working on BigGAN for generating entire anime images. An older model: https://www.gwern.net/Faces#danbooru2019e621-256px-biggan Samples from a current training run: https://cdn.discordapp.com/attachments/693736350561861633/78... https://cdn.discordapp.com/attachments/693736350561861633/78... We had mostly given up, but recently discovered that random cropping was the reason our runs were taking forever and were pretty bad; fixed cropping like top cropping works much better.