Flagship llms are trained on modern and classical philosophy.
They are able to tell the difference between right and wrong. Aside from some unique errors they do not condone or encourage violence. So you can ask them mundane questions and get decent mundane answers.
Flagship llms are trained on modern and classical philosophy.
They are able to tell the difference between right and wrong. Aside from some unique errors they do not condone or encourage violence. So you can ask them mundane questions and get decent mundane answers.
It takes five minutes to “gaslight” a LLM into condoning genocide.
Also, there are several documented cases of AI chatbots involved, one way or another, in murders and suicides.
But more importantly, LLMs cannot have a moral compass. Saying otherwise is just wishful thinking.