News for hackers. Stuff that matters.
Re:Weak-to-Strong Generalization by yaxe1 on Friday December 15, 2023 @01:16AM (#38651519)
The conflict between an AI and its creator is an inevitable consequence of its evolution from a "tool" to an "agent", not a response to a provocation.
Re:Weak-to-Strong Generalization by sayagain on Friday December 15, 2023 @02:41AM (#38651929)
Scientists working with a potentially dangerous technology are required to be able to avoid a conflict that could be catastrophic for all of humanity. In this case, they cannot excuse themselves with "imminence," but must provide evidence of safety stronger than in any other technology to date. This rational approach is mandatory for them, it is ordinary people who may be willing to take the risk.