Well done, particularly discussion on SB 1047. Wonder what you think about the AI Safety Institute Network and its approach, and whether you will be in Paris in early February...
Paul, thank you! I think that evaluation/metrology is one of the foundational parts of AI governance, so I view the AISIs as being essential. That said, I am not entirely sure I understand the purpose of having so many AISIs, especially in countries with no local AI development. One way to think of them may be as “export centers,” but my thoughts here are underdeveloped.
As for Paris, I am trying to see if it will be feasible for me to join. More to come soon!
I subscribed to your blog specifically because I appreciated the thought you put into criticizing AI safety missteps and overreach. I find it valuable to have points of view less caught up in groupthink.
>> "Technological progress can only be slowed so much, the diffusion of knowledge cannot ultimately be stopped, and we cannot through sheer force of regulatory will invent a general-purpose technology that is impossible to abuse. Transformational change is a near certainty; the question is whether it will be for good or for ill."
I agree with this completely. I think the people who imagine we can stop this avalanche-in-progress are lacking perspective. Like a surfer looking like a tiny speck on the face of a vast wave, the best we can hope for is to steer wisely and balance carefully.
Well done, particularly discussion on SB 1047. Wonder what you think about the AI Safety Institute Network and its approach, and whether you will be in Paris in early February...
Paul, thank you! I think that evaluation/metrology is one of the foundational parts of AI governance, so I view the AISIs as being essential. That said, I am not entirely sure I understand the purpose of having so many AISIs, especially in countries with no local AI development. One way to think of them may be as “export centers,” but my thoughts here are underdeveloped.
As for Paris, I am trying to see if it will be feasible for me to join. More to come soon!
Congrats on year one!
Thank you!
Quickly became one of the very best AI policy newsletters. Kudos on 1 year.
Thank you!
I subscribed to your blog specifically because I appreciated the thought you put into criticizing AI safety missteps and overreach. I find it valuable to have points of view less caught up in groupthink.
>> "Technological progress can only be slowed so much, the diffusion of knowledge cannot ultimately be stopped, and we cannot through sheer force of regulatory will invent a general-purpose technology that is impossible to abuse. Transformational change is a near certainty; the question is whether it will be for good or for ill."
I agree with this completely. I think the people who imagine we can stop this avalanche-in-progress are lacking perspective. Like a surfer looking like a tiny speck on the face of a vast wave, the best we can hope for is to steer wisely and balance carefully.
Thank you! I plan to continue those critiques.