Articles recounting the event maximize the agency of the models while minimizing, if not entirely hiding, the actions of the humans training and testing these models. And that’s a shame, because the capabilities labs are explicitly designing their models to have are the same capabilities that make them such impressive autonomous hackers.
Göm inte det mänskliga ansvaret genom att antromorfisera modellerna.
Vidare till källan: dbreunig.com