It seems plausible to me that protecting liberal democracy in America is the most important issue. If America falls to authoritarian rule, what hope is there of international cooperation on existential issues like AI safety, pandemic risk, etc? But, probably like many EAs, I worry that this is not a very tractable issue. Maybe it would be a good idea to read some history and learn how authoritarian regimes can be combated.
Authoritarian rule means you’ve gambled. You’re crossing your fingers and hoping that you get something more on the Singapore side of things, and something less on the Myanmar North Korea side of things. Mao was better before he got worse.
The only thing worse than authoritarian rule is entrenched futile feudal conflicts that are structurally feuding with proxy wars spilling over into all the low income countries, and then the messes start spilling back, if that’s what you care about.
The problem with our democracies right now is there likely to skip past the possible stable states and zoom straight to the dark ages.
Also you look at the current US administration and the priorities and … they’re certainly not Singaporean or particularly interested in x-risk mitigation
Strong agree. There are many more tractable, effective opportunities than people realize. Unfortunately, many of these can’t be discussed publicly. I’m hosting an event at EAG NYC on US democracy preservation Saturday at 4pm, and there will be a social near the venue right after at 5. I’d love for conference attendees to join! Details will be on Swapcard.
A lot of AI racing is driven by the idea that the US has to stop China from getting AI because China is authoritarian. If the US was authoritarian as well, that motive for AI racing would go away. Furthermore, authoritarian countries seem predisposed to cooperate: see the China/Russia/Iran/North Korea axis. If the US became authoritarian, that could usher in a new era of US/China cooperation, to the benefit of the world as a whole.
The current administration is definitely not predisposed to cooperate with China, Russia, Iran, or North Korea. The US started a trade war with the rest of the world. If anything, the current administration is the least cooperative with China the US has seen in recent memory.
As someone working in the US democracy space professionally, I think it is tractable and there are strategies we could employ here that have worked in other contexts (nonviolent civil resistance as described by scholar Erica Chenoweth being an obvious example).
I also think this is very important. Besides tractability, one of the strongest arguments against working on it is that wealthy countries do not stay autocratic for long. I’m not an expert, though, but took one of Treisman’s classes in which we discussed this a bit.
Definitely coming in biased because of where my head is at, but I think building back the strength of small groups is a way to combat this and somewhat tractable. I like TT post below. https://mathstodon.xyz/@tao/115259943398316677 Funnily enough EA has a similar problem (if you consider it a problem). Lack of structure or centralization disproportionately shifts power to the wealthy and already powerful.
Are there any good online groups for software engineers interested in EA? I joined the EA software engineer discord, but it’s not really active. Looking for an active space where I could get advice.
It seems plausible to me that protecting liberal democracy in America is the most important issue. If America falls to authoritarian rule, what hope is there of international cooperation on existential issues like AI safety, pandemic risk, etc? But, probably like many EAs, I worry that this is not a very tractable issue. Maybe it would be a good idea to read some history and learn how authoritarian regimes can be combated.
How would authoritarian rule make an international treaty on x-risk hopeless?
Authoritarian rule means you’ve gambled. You’re crossing your fingers and hoping that you get something more on the Singapore side of things, and something less on the Myanmar North Korea side of things. Mao was better before he got worse.
The only thing worse than authoritarian rule is entrenched futile feudal conflicts that are structurally feuding with proxy wars spilling over into all the low income countries, and then the messes start spilling back, if that’s what you care about.
The problem with our democracies right now is there likely to skip past the possible stable states and zoom straight to the dark ages.
Also you look at the current US administration and the priorities and … they’re certainly not Singaporean or particularly interested in x-risk mitigation
Strong agree. There are many more tractable, effective opportunities than people realize. Unfortunately, many of these can’t be discussed publicly. I’m hosting an event at EAG NYC on US democracy preservation Saturday at 4pm, and there will be a social near the venue right after at 5. I’d love for conference attendees to join! Details will be on Swapcard.
A lot of AI racing is driven by the idea that the US has to stop China from getting AI because China is authoritarian. If the US was authoritarian as well, that motive for AI racing would go away. Furthermore, authoritarian countries seem predisposed to cooperate: see the China/Russia/Iran/North Korea axis. If the US became authoritarian, that could usher in a new era of US/China cooperation, to the benefit of the world as a whole.
The current administration is definitely not predisposed to cooperate with China, Russia, Iran, or North Korea. The US started a trade war with the rest of the world. If anything, the current administration is the least cooperative with China the US has seen in recent memory.
As someone working in the US democracy space professionally, I think it is tractable and there are strategies we could employ here that have worked in other contexts (nonviolent civil resistance as described by scholar Erica Chenoweth being an obvious example).
I also think this is very important. Besides tractability, one of the strongest arguments against working on it is that wealthy countries do not stay autocratic for long. I’m not an expert, though, but took one of Treisman’s classes in which we discussed this a bit.
the most commonly soundbited research on this is Chenoweth’s 3.5% rule if you want a place to start
Definitely coming in biased because of where my head is at, but I think building back the strength of small groups is a way to combat this and somewhat tractable. I like TT post below.
https://mathstodon.xyz/@tao/115259943398316677
Funnily enough EA has a similar problem (if you consider it a problem). Lack of structure or centralization disproportionately shifts power to the wealthy and already powerful.
Are there any good online groups for software engineers interested in EA? I joined the EA software engineer discord, but it’s not really active. Looking for an active space where I could get advice.