Forum Replies Created
-
Posts
-
I may be confused by definitions. Taken literally, “maximize pleasure” would prioritize pleasure above all other values; this is actually an example used as a cautionary tale for the difficulty of correctly programming a superintelligence (program your superpowerful AI to literally maximize human pleasure and you end up with brains being grown in vats having their pleasure centers electrically stimulated).
Thus I would have been surprised if any effective altruists had said that they had a goal of maximizing pleasure, which is why I asked for a citation.
I don’t think that minimizing suffering and maximizing pleasure are opposites of the same scale; I think people can care a lot about not inflicting unnecessary suffering on others without also having pleasure as a primary or dominant value.
Regarding the Apollo and Dionysus essay contrasting the people watching the Apollo moon landing and the people attending the Woodstock music festival, values such as caring about children on the other side of the world that you’ll never meet not dying of malaria strikes me personally as rather more of an Apollonian kind of value than an Dionysian one.
I wouldn’t be surprised to see the effective altruists I know watching a moon landing; for many of them I would be surprised to see them somewhere like Woodstock.
I would say that my observation is that many effective alturists are high in the personality trait of openness to experience as described in the Big Five personality model: as described on Wikipedia, having an active imagination (fantasy), aesthetic sensitivity, attentiveness to inner feelings, preference for variety (adventurousness), intellectual curiosity, and challenging authority (psychological liberalism).
Openness to experience is strongly anticorrelated with political conservatism; I don’t know offhand of any effective altruists who are conservatives. I think this is a different dimension than Apollonian-Dionysian; e.g. there are Dionysian angry Trump base type conservatives and Dionysian radical left Woke types; there are both liberal and conservative Apollonians who care about truth and the “dedication to the absolutism of reality”.
I think the openness of many effective altruists to unconventional ideas and lifesyles such as polyamory is more related to the openness to experience personality trait than Dionysian values. Effective altruists care more about the truth than anyone else I know.
I don’t know anything about apocalypticism aside from what I’ve read on Wikipedia, but to the extent that Wikipedia is correct that apocalypticism represents a belief that the end of the world is imminent, I don’t think that accurately describes effective altruists. Effective altruists are concerned about existential risks and seek to avert them; this is an Apollonian project in the same way that it’s an Apollonian engineering project to build a moon rocket that doesn’t explode and kill the astronauts.
Oh my. I had completely missed that Altman had signed the “Mitigating the risk of extinction from AI should be a global priority” statement back in May. I had been under the mistaken impression that Altman was in the “totally 0% chance of existential risk” camp. In that context Altman’s tweet struck me as a bit unkind, and I don’t think Altman is an unkind person, so I had been reaching for an explanation. But given that the disagreement between Altman and Yudkowsky is narrower (e.g. Yudkowsky thinks there’s at least a chance that current AI technology could be existentially dangerous, while Altman doesn’t think so), Altman’s tweet no longer strikes me needing explanation. So I retract my speculation for that part.
Yes, I agree with your analysis of the intent of the manifesto. I mean, I’m a technologist, and generally optimistic about technology, and I found the manifesto repugnant… so it certainly failed to inspire me 🙃. But I expect that my dislike for the manifesto colored my impression of its purpose.
Right, the answer to “how many years later” might be “never”! Perhaps in the counter-factual scenario China would have ended up capturing the low-cost medium/heavy lift launch market, and US companies would have found the barrier to entry too high to start competing, and too hard to independently develop the necessary technical skills . An industry might be stifled by government regulation, a country might go into economic decline, perhaps by the time a technology could have been developed it would have been obsolete.
“Uber for X” is a bit different actually, at least when used correctly. A new market, not having been done before, can be lengthy to explain. “Uber for X” is intended to be an analogy to give a quick idea of what it’s about… which if might be of interest, could then lead to the lengthy explanation. Of course, there are also the copycats and ideas that turn out to be duds.
Even Rationalism/EA, filled with nerdy Apollonians, has as its goal the
Dionysian minimization of suffering and the maximization of pleasure.Citation?
The way I think about it is, how many years did the presence of a person accelerate technological development? For example, in 2012 the United States had 87 rocket launches, of which 60 were Falcon 9. Without Musk and SpaceX, how much longer would it have taken the United States to reach 87 rocket launches a year?
For any particular technology, we start out with it being impossible to do, and then can only be done by nation-states, and then eventually it can be done by super motivated hardcore billionaire genius entrepreneurs, then it becomes easy enough that ordinary companies can do it, and sometimes we get to the point where small groups or individuals or small groups can do it.
So today we’re at the point where ordinary companies can develop commercial lift rocket vehicles, and we have various commercial competitors to SpaceX finding niches in the market. Back in 2008, for SpaceX’s first commercial flight, only Musk could do it. The Electron rocket is smaller than the Falcon 9, so this isn’t really a correct comparison, but it had 9 launches last year while SpaceX had 8 launches in 2016, so that’s a minimum of a 6 year acceleration.
Of course we’d need to add more years for someone to develop a rocket as large as the Falcon 9. Counter-factually, if SpaceX hadn’t existed, other entrepreneurs might have entered the market sooner. SpaceX quickly captured the market by bringing costs so low, and that made it harder for competitors to get started. But I can certainly imagine that Musk provided at least a 6 or 8 year acceleration.
What about Altman and OpenAI? The current hot AI technology, large language models (LLM), takes enormous capital to train. So it was something that Google, Facebook, and Microsoft were working on (and something that only large companies could have worked on), but they weren’t working on it very urgently. OpenAI lit a fire under everyone, e.g. Google management declared a “code red” in response to the competitive threat suddenly posed by ChatGPT. So I imagine that Altman might have accelerated LLM development by perhaps 2 years, at a guess?
At this point, we don’t know whether LLM’s in particular are dangerous in terms of an AI apocalypse or not. They could be, in which case developing capabilities would be a bad idea, or they might not be, in which case we’d be giving up the potential benefits unnecessarily if we pulled back. So if Altman accelerated LLM development by 2 years, that doesn’t necessarily means that we’re 2 years closer to apocalypse.
As for LessWrong, the counter-factual is if LessWrong didn’t exist, how much longer would it have taken Altman to get interested in AI? Here’s where I’m dubious about the claim that LessWrong made much of a difference. There were many sources of information about AI, and LessWrong happened to be the one that Altman was reading.
Regarding tribal allegiances to tech billionaires, it’s interesting to read Andreessen’s “Techno-Optimist Manifesto” from the viewpoint of Paul Graham’s essay “The Four Quadrants of Conformism” (http://paulgraham.com/conformism.html) about passive vs. aggressive, conventional-minded vs. independent-minded people.
Andreessen’s “Manifesto” is anathema to someone independent-minded, consisting of a long list of what we’re supposed to believe and what we’re supposed to not believe. So it’s directed to the conventional-minded, in an attempt to convert such people to believing what he wants. Yet Andreessen himself is of course independent-minded. So it’s a bit like a cult leader in that way, a prescription for others that the leader wouldn’t do himself.
Lessons about drones: https://spectrum.ieee.org/ukraine-drone-war
How realistic would it be for the US to resupply Taiwan if China were to blockade or invade?
I’m wondering if the “Oh whoops they got invaded, let’s quickly ship them 17,000 antitank weapons” approach might not work as well without the hundred of miles of land border with a friendly country?
A blog post explaining SWIFT by Patrick McKenzie (patio11): https://bam.kalzumeus.com/archive/moving-money-internationally/