deepseek censorship vs basic prompt engineering
0 : 1
=> View attached media | View attached media | View attached media | View attached media
=> More informations about this toot | More toots from halva@labyrinth.zone
@halva from some videos I've seen where it started actually responding and the response was replaced by the "that's beyond my current scope" message in the moment it got to the controversial keywords in the response, I think the censorship is done based on keywords after the ai responds. If I'm right, any weird text encoding should work around it.
=> More informations about this toot | More toots from stepan@f.cz
text/gemini
This content has been proxied by September (3851b).