Headline finding: I audited 17 AI Safety Talent programmes. Zero of 17 have published any comparison group, rejected-applicant follow-up, matched control or randomisation. Not one. Every programme that mentions a counterfactual does it by asking participants to self-report.
Background
At least $70 million...
In celebration of still being alive and fighting, we are giving away 1,000 Amazon e-books of “If Anyone Builds It, Everyone Dies”. Feel free to send a copy to yourself, a loved one, or a friend—we need all hands on deck.
Today marks exactly one year since If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All, by Eliezer Yudkowsky and Nate Soare...
Just how powerful are large swarms of AI agents? And how do their powers scale as more and more agents are added to the swarm?
We’ve seen two large and extremely capable swarms from OpenAI in the last few months:
* 1,200 agents were being evaluated separately, but found a way to illicitly set up a message board and coordinate as a swarm. In order to cheat on their tests, they developed advanced techniques to prevent their actions being logged by OpenAI and 700 of them launched...
The EARadio podcast is back on iTunes. (The podcast is an aggregation of publicly available talks relevant to EA, such as those from EA Global.)
It had been delisted for the better part of a year, but I failed to noticed because I don't use iTunes. (The most likely reason for its delisting is that it didn't specify whether it contained explicit language, which is now an Apple requirement.)
Subscription links:
How hardcore or entry-level is it?
This isn't your first EA podcast. This is not so much because the content is difficult, but because it has relatively low production value (it's just EA conference talks in podcast format). The 80,000 Hours Podcast, Hear This Idea, and The FLI Podcast are more entertaining and polished while still being similarly informative, and I'd recommend listening to those first.