The fundamental problem with SB 1047 was that it regulated AI models rather than AI systems. "Is this model dangerous" is an impossible question to answer because it depends on how the model is used. understandingai.org/p/governor-new
5,628
ViewsPost
Conversation
![David Watson 🥑](data:image/svg+xml,<svg xmlns="http://www.w3.org/2000/svg" width="96" height="96"><rect fill-opacity="0"/></svg>)
Post your reply
Re: "it’s not reasonable to expect the creator of a model to anticipate and prevent all harmful uses." It sounds as though you are saying that open weight machine intelligence models are intrinsically dangerous. That phrasing might be more acceptable to the doomers.
An open-weight model is intrinsically dangerous in the same sense that a kitchen knife or a bottle of bleach are.
their liability requirements were the real ptomaine in their stew: who in their right mind would fine Toyota every time a Prius runs a stop sign?
All the best AI models in one place. For one price.
I mean... it makes sense to me IDK.
I think this highlights the need for more nuanced policy discussions around AI. Regulating systems rather than models requires a deeper understanding of how AI is actually used.
I completely agree, regulating models is too narrow, we need to focus on how these models are integrated into systems and used in real-world applications.
Show additional replies, including those that may contain offensive content