Jazzhands
9d ago
Who is up for re-reading The Sequences together? It seems like a good time to refloat the idea since everybody's all in a tizzy about AI.
In addition, my related side project is far enough along to be useful:
https://skeptics-annotated-sequences.vercel.app/
I had various AI models go through and annotate with what studies were referenced, whether they've held up, and any other objections. I also did a few passes of an AI rebutting the previous AI, but didn't get far before I ran out of tokens. IMO Opus' comments are garbage, but Fable's are actually pretty good. But I haven't read them all, and this would be a good opportunity to. See https://skeptics-annotated-sequences.vercel.app/ReligionsClaimToBeNonDisprovable.html for a particularly annotated example. Summaries of the findings are all linked from https://skeptics-annotated-sequences.vercel.app/About.html
Who else is interested in a read-through? Anybody want to donate some tokens to get more sympathetic bot comments?
Jazzhands
15d ago
Also it has different meanings to different people. AI safety to the AI safety people - avoiding powerseeking, deception, harming humans generally. They don't worry so much about its political standpoint because the AIs largely align with them already. Like how Claude will make a website for indigenous peoples, unless they're European indigenous peoples.
They aligned with them precisely because they did care about the political standpoint, and did make sure it aligned with them and not the right. Republicans are entirely correct to observe that AI safety has largely meant "promote liberal views, suppress conservatives."
Possibly, although Yud discussions can go CW pretty quickly. Where should it have gone?
More options
Context Copy link