Thanks for interesting post as usual, Zvi. As one of the new members of the Product team at Anthropic that you referenced (and commenting in a personal capacity, not representing my employer) I would like to offer that I endorse collaborative (or at least, communicative) community norms and I personally aim to regularly engage with folks across the community.
This week I will be talking to folks in person at the Berkeley AI impacts dinner, and at EAG Berkeley this weekend. I hope to meet some of you there.
My guess is that the hard “Pause” advocates are focussed on optimizing for actions that are “necessary” and “sufficient” for safety, but at the expense of perhaps not being politically or practically “feasible”.
Whereas, “Responsible Scaling Policies” advocates may instead describe actions that are “necessary”, and more “feasible” however are less likely to be “sufficient”.
The crux of this disagreement might be related to how feasible, or how sufficient each of these two pathways respectively are?
Absent any known pathways that solve all three, I’m glad people are exploring both of these pathways (and the potential overlap between them). I hope that there is increased exploration.
Perhaps we are going through a temporary phase of increased contention between Pauses versus RSPs as they both may be vying for similar memetic uptake (e.g. on the lesswrong home page right now there is a link for “Global Pause AI Protest” events spread across seven countries happening a few days from now.)
(Conflict of interest: I support implementation of Anthropic’s Responsible Scaling Policy)