Member of the LessWrong 2.0 team.
I see what you're getting at. Interesting question.
Curated. This is a great instance of someone increasing clarity on a load-bearing concept (at least in some models) through an earnest attempt to improve their own understanding.
My understanding is that you can't safely do even A with an arbitrarily powerful optimizer. An arbitrarily powerful optimizer who's reward function is solely "beat the grandmaster" would do everything possible to ensure it's reward function is maximised with the highest probability. For instance, it might amass as much compute as possible to ensure that it's made no errors at all, it might armor it's servers to ensure no one switches it off, and of course, it might pharmacologically mess with the grandmaster to inhibit their performance.
The fact that it can be done safely by a weak AI isn't to say that it's safe to do with a powerful AI.
Curated. I don't think I can say it better than Habryka did in a past year:
This post has been an amazing resource for almost anyone who is working on AI Alignment every year. It also turns out to be one of the most comprehensive bibliographies for AI Alignment research (next to the AI Alignment Newsletter spreadsheet), which in itself is a various resource that I've made use of myself a few times.
Thank you very much for writing this!
You could try using Greaterwrong.com and see if they're viewer is easier to find your place again. I'll see what else I can think of.
Curated. Although written a while ago, I love posts discussing how to approach figuring out the world. I'd be interested to see more examples of the process advocated here though. Where are the cases of this working?
My concern is that without the brackets, they're even smaller and harder to see and click.
Frontpaging this despite the current policy to leave most Covid content on Personal Blog. Doing so because it is of interest and broad relevance to people's decisions and seems high-effort and high-quality, though note questioning of biased reasoning in the comments.
Jacob Lagerros got excited about RaDVaC early and came to the Lightcone team a few months ago saying we should arrange funding for them. I believe talks have occurred between relevant parties though I'm not certain of exactly where things ended up.
Summary: I suspect funding has been provided, mostly want to give credit to those who noticed early and worked to make it happen.
Curated. For most of the Covid pandemic, I've heard Rationalists say that we should have (or to have) invested in making our vaccines given how slow release of the mainstream ones was. It's very cool that caffemacchiavelli actually worked on this! I wasn't aware of the project until now, so major kudos. I hope going forward more people follow suite. Also, cool to know that this project was inspired by other LessWrong posts–fantastic to see them having that impact.