This is really thought provoking work. I think the power of personal diplomacy may have been overplayed at the point where ‘the President’ and Xi are mentioned hashing out capabilities limitations over phone call. I understand it’s an optimistic scenario, but these things are generally procedural. Is that a baked in assumption?-- that the standard means of accomplishing international cooperation will have to fall by the wayside in favor of a more gung-ho, personal style of diplomacy if something like The Coalition is to be achieved? I’m generally interested in how you guys made your assumptions re: where incentive structures could carry the day vs. where more concerted political will would have to be built.
Carl Parkin
Karma: 4
Instead they worked hard, and collected a coalition, and built an international nuclear anti-proliferation regime.
I think it’s important to note that this happened after two catastrophes—Hiroshima and the Cuban Missile Crisis. I think the overall sentiment of this article is probably correct, and law is necessary. The question is how it can come about. The ‘working hard’ bit here deserves more elaboration, especially regarding the necessary conditions for hard work to actually pay off. Is a catastrophe required for AI to get its non-proliferation regime?
This is a great point. AI safety spawned from tech communities, and a lot of the people in the field have backgrounds in CS, etc. because that’s what predisposes you toward understanding the risk. But that social background / group is inherently going to valorize technical work, research, etc. over talking to people and trying to change minds, because that’s the way tech fields have seemingly worked for the past several decades. The most common story about tech is ‘if you build it they will come’—if you invent the computer or the smartphone, you also invent the market for the computer or the smartphone, because people just realize that it will improve their lives in crazy ways and want one.
But there are marketing budgets behind these things. The sad truth is that empirics need to be socially leveraged to be made meaningful. Technology diffuses socially, it’s not a story of a million individual realizations that the IPhone will marginally improve their happiness/productivity, but a sprawling and expanding network of people who buy it because their friend bought it and likes it. The same is true for AI safety. Yes, the empirics existing is important, but just the empirics existing isn’t going to cause awareness to spread.
Maybe I’m oversimplifying the view of tech folks, but that’s how they often seem to me.