The inordinately slow spread of good AGI conversations in ML
| Owners | ||
|---|---|---|
| Reserved | 0 | – |
The article: “The inordinately slow spread of good AGI conversations in ML”
Fundamental Justification
AGI risk is very high. Two key causes for this are "not much conversation about AGI is happening within ML", and "the conversations that do happen are relatively low-quality and non-candid". Making EA and ML more aware of these dynamics increases the odds that they'll be spontaneously solved to some degree (because individuals can consciously change their behavior now that they see the problem), and also enables EAs and ML researchers to come up with future deliberate strategies to try to solve this problem.
What outcomes were to be expected before you started the project? What were unusually good and unusually bad possible outcomes? (Please avoid hindsight bias and take the interests of all sentient beings into account.)
In general, a given blog post is unlikely to have much impact. Most attempts like this will fail; occasionally, one may succeed dramatically, as illustrated by the Elon Musk example in the post (https://www.lesswrong.com/posts/Rkxj7TFxhbm59AKJh/the-inordinately-slow-spread-of-good-agi-conversations-in-ml) and by the COVID-19 example in https://www.lesswrong.com/posts/Js34Ez9nrDeJCTYQL/politics-is-way-too-meta.
An example of an unusually bad outcome would be if lots of ML researchers see this post and are offended because they feel straw-manned by it.
What actual outcomes are you aware of?
Not much, just a bit of engagement here, on Twitter, and on LW (https://www.lesswrong.com/posts/Rkxj7TFxhbm59AKJh/the-inordinately-slow-spread-of-good-agi-conversations-in-ml).
Who can make a legitimate claim to a fraction of the impact, and have you talked to them?
The people who have shaped my background views on AI the most are Eliezer Yudkowsky and Nate Soares.
Who are the current owners of the impact and what fraction do they each own?
I own all of the impact.
Procedural Questions
What is your minimum valuation under which you’ll not sell any shares in your impact?
I wouldn't sell my full impact for less than $200.
What would you have done had there been no chance to get retro funding? (This helps us assess our impact but has no effect on our evaluation of the certificate’s impact.)
In this case, I would have done exactly the same series of things. It wasn't on my radar that I might get funding.
What can we improve about this process?
Make it shorter and simpler.