deepfates

|||

Signal GAN

I trained a GAN on SIGNAL

In 2021 I used the raw images from SIGNAL to train a generative adversarial network, or GAN, to generate in-between images of the pages. It was one of my first experiments with GANs and wasn’t very good, but I posted a tweet with a video of the training process.

That tweet went semi-viral: over 1,200 likes and hundreds of retweets, huge for roughly 800-follower account at the time. Many people accused me of summoning cyberdemons”, which is apparently a thing from Buffy the Vampire Slayer and also maybe the video game DOOM.

This proved my memetic engineering thesis to my satisfaction, but it also drew unwanted attention from Twitter users and… other entities. So I deleted it.

The following sample has been selected for its relative safety. Nonetheless, please exercise caution when viewing it, especially for long periods.


Up next Autocomplete everywhere The Big Sleep Previously, I wrote about auto-completing language models — Transformers — and the feedback loop they create with our minds: Robert Anton Wilson, a
Latest posts When will a human level AI be built? Silicon AI Sparkl Invisible Movies The Mirror of Language Worldview interview Memery Circles Language Arts series Language Arts 1.0 Language Arts 0.3 Language Arts 0.2 Language Arts 0.1 The Big Sleep Signal GAN Autocomplete everywhere Learning loops Seasons change Berduck Book Cover Generator See and Point New social gestures Digital gardens Zones of the mind Legiblate Where the humans end Accelerating succession Gardening algorithms The law of the instrument A cybernetic meadow Robot Face