Comment by singiamtel

I found this principle particularly interesting:

    Human oversight: The use of AI must always remain under human control. Its functioning and outputs must be consistently and critically assessed and validated by a human.