4 ms·
Uh, because the model started spewing virulent hate speech a few days ago? What normal software does this?
by archagon 1y ago
Uh, because the model started spewing virulent hate speech a few days ago? What normal software does this?
- tordrt 1y agoNot the model itself, the X bot. Its obvious that this has happened due to them tweaking the bot, you could never get it to write anything like this a couple of weeks ago.
- hengistbury 1y agoCan you trust the model when the people releasing it are using it in this way? Can you trust that they won't be training models to behave in the way that they are prompting the existing models to behave?
- block_dagger 1y agoAny LLM trained appropriately. Tokens in, tokens out.
- Levitz 1y agoIt wasn't that long ago that we had "normal software" turning everybody black. This is just how AI works, we humanize it so it's prone to controversy.
- briangriffinfan 1y agoAn acute memory will remember this happening with basically every chatbot trained on text scraped from the internet, before they had to explicitly program them to avoid doing that.