In my experience it's worse than that's, it's easy to set off by using any of a number of trigger words.

Many uses of the word "black" for example (even if you are just talking about a black notebook) make it start using racial stereotypes.

> How do you prevent AI being an A grade jerk.

Invent synthetic consciousness and ask it to be nice, easy :) I'm only half joking, we probably all have thoughts ranging from bad to horrible, but we just don't say them because we are aware of the consequences. Language models aren't aware so they'll spit out the most likely combination of words. If there would be a process to limit these or try again, it could act as a filter, but I think that requires it to be self aware.

This is why we've built security policies at Mantium. You can run the input and output through an offensive speech detector, and halt replies the prompt if "badness" is detected. This is, of course, an imperfect system because philosophies around what is offensive can be very diverse, but we find that security policies are helpful.

Maybe their mama didn't raise it right? They should raise it on information from good people. Now, how to find those "good people comments"?