How close are current AI agents to automating AI R&D? Our new ML research engineering benchmark (RE-Bench) addresses this question by directly comparing frontier models such as Claude 3.5 Sonnet and o1-preview with 50+ human experts on 7 challenging research engineering tasks.
It's too bad "effective altruism" has been destroyed as a brand by SBF, because altruism is important and "effective" is probably the most important adjective you could qualify "altruism" with.
As the head of the Center for Effective Altruism, I was dismayed to learn that some EAs go by their initials. I am hereby requiring everyone who cares about their impact to adopt a style of first name, last initial, all lowercase.
