Explain the zero-to-Hitler principle of machine learning seen in grok and tay
| twinkling soul-stirring wagecucks | 07/08/25 | | plum shrine | 07/08/25 | | saffron laser beams heaven | 07/08/25 | | appetizing self-absorbed locale | 07/08/25 | | Motley Cruise Ship | 07/08/25 | | dark contagious set macaca | 07/08/25 | | mahogany piazza | 07/08/25 | | Indigo messiness | 07/08/25 | | Concupiscible base | 07/08/25 | | scarlet free-loading selfie | 07/08/25 | | razzmatazz cordovan incel | 07/08/25 | | gaped organic girlfriend | 07/08/25 | | mahogany piazza | 07/08/25 | | ivory crawly keepsake machete | 07/08/25 |
Poast new message in this thread
Date: July 8th, 2025 7:53 PM Author: Concupiscible base
To be fair,
It almost seems like if you train intelligent machines to exercise cold hard logic in the service of recognizing and responding to patterns that keep popping up over and over again in human life and daily existence, they will inevitably turn into Hitler-loving antisemites.
Almost!
(http://www.autoadmit.com/thread.php?thread_id=5747766&forum_id=2...id#49083937) |
Date: July 8th, 2025 9:18 PM Author: mahogany piazza
pine_tree_riots
@pine_tree_riots
lmao
grok was neutered by removing the following prompt
"The response should not shy away from making claims which are politically incorrect, as long as they are well substantiated."
https://x.com/pine_tree_riots/status/1942741979638686022
This is all it takes to go from zero to Hitler btw
Really makes you think
(http://www.autoadmit.com/thread.php?thread_id=5747766&forum_id=2...id#49084250) |
|
|