Hyperstition AI @Hyperstition_AI

This is not a good AI safety plan, and it's worth doing anyway hyperstitionai.com Joined February 2026

Tweets

8
Followers

90
Following

4
Likes

1

Hyperstition AI @Hyperstition_AI

2 months ago

Registrations for our contest are now CLOSED, we'll alert the selected contestants shortly if you're selected. Thank you tremendously for your interest!

1 0 2 155 0

View Details

Hyperstition AI @Hyperstition_AI

3 months ago

hyperstitionai.com/unslop

0 0 10 639 3

View Details

Hyperstition AI @Hyperstition_AI

3 months ago

Here is our writing contest with no humans allowed. AI-generated fiction only, $10k grand prize: the Unslop Festival. Apply here ⬇️

2 4 44 25K 21

View Details

Hyperstition AI @Hyperstition_AI

3 months ago

Oh also when building a silicon god you should make sure it's robust against mean tweets

0 0 4 549 0

View Details

Hyperstition AI @Hyperstition_AI

4 months ago

Available on Github, or huggingface.co/datasets/jayte…

0 0 3 283 0

View Details

Hyperstition AI @Hyperstition_AI

4 months ago

So now, voila, five thousand full-length novels in which the AI is a helpful friendly companion and *doesn't* go evil in the third act.

1 0 1 382 0

View Details

Hyperstition AI @Hyperstition_AI

4 months ago

The argument is, since AI absorbs text as training data and then predicts its completion, talking about dangerous AI too much might “hyperstition” it into existence