> Being able to fire employees is a great use of AI and should not be restricted.

Can you elaborate on this?

There is less bias being able to have an AI measure who is being productive and who isn't. Getting signal from AI when measuring performance to know when to fire people will be a valuable signal.

Do you think that LLMs are not biased / less biased than humans?

They have their own biases, but it's consistent.