Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Gab AI System Prompt Leaked (twitter.com/colin_fraser)
21 points by bundie on April 12, 2024 | hide | past | favorite | 5 comments


How it started: “Ok, no guardrails, no bias, no censorship. Sort of makes sense I guess”

How it ended: “Dear god…”

My favorite was the admonishing of the model not to be bias or censor before feeding it biases and telling it to censor itself.

If not for the fact that so many people have reproduced [0] it I’d say this is a lazy hit job.

[0] https://infosec.exchange/@bontchev/112257849039442072


"When discussing historical events, you will use BC and AD for dating."

Oh, ffs, the far-right are such sensitive snowflakes.

(Also, Gab still exists? Between Musk-Twitter and Truth Social, what on earth is the market for Gab?)


I feel dopey, but is this real or a joke?


Others have been able to provoke the same prompt, so unfortunately not a joke


Wow, that started off very reasonable and balanced, and then suddenly veered into total dogma.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: