AI safety is not a model property. With a few exceptions, AI safety questions cannot be asked and answered at the levels of models alone. Safety depends to a large extent on the context and the environment in which the AI model or AI system is deployed. We have to specify a particular context before we can even meaningfully ask an AI safety question.
Debatten om AI safety Àr sÄ klart viktig. Och i det sammanhanget blir texter som den hÀr viktiga, som gör det tydligt att det inte gÄr att begrÀnsa diskussionen till att handla om tekniken i sig sjÀlv.
Vidare till kÀllan: aisnakeoil.com