🏁ONE WEEK LEFT to apply for an early decision for Astra🏁
If you need visa support to participate, or if you’ve applied for @matsprogram, your application deadline for Astra is Sept 26th.
⬇️We're also excited to announce new mentors across every stream!
(1/4)
🚀 Applications now open: Constellation's Astra Fellowship 🚀
We're relaunching Astra — a 3-6 month fellowship to accelerate AI safety research & careers.
Alumni @eli_lifland & Romeo Dean co-authored AI 2027 and co-founded @AI_Futures_ with their Astra mentor @DKokotajlo!
🧵 1/8 I’m building a world-class team of Research Program Managers at @ConstellationAI to advance AI safety! 🚀
Looking for experienced operators who can design and build the research ecosystem needed to safely navigate transformative AI.
Application in thread!
@T_Jaenichen There are definitely reasonable critiques to be made, and the risk that this have a net-negative effect on safety is among them. But it still seems premature - the outcomes remain to be seem, and the action is hard to judge ex ante absent full context.
Back to twitter to read on OpenAI news and wow! Seems so bad to base any takes on press leaks esp from anon sources in the middle of this. Many ppl “familiar with discussions” talking to press trying hard to swing public and internal opinion, likely super misleading/biased info.
Announcing two Constellation programs! Apps due Nov 10
Astra fellowship: do research with mentors like @EthanJPerez@OwainEvans_UK@RichardMCNgo@ajeya_cotra https://t.co/NGQWPKyIBY
Visiting researchers: exchange ideas with people from top AI safety orgs https://t.co/dXcWKFMkp3
@robertwiblin Humans also take 9 months to reproduce, are very noisy while doing it, and then at least 9 more years before the offspring is even somewhat mature.
A capable AI could silently and invisibly reproduce fully capable offspring in about 9 minutes.
Adderall and marijuana: wrecking dopaminergic pathways; neuropsychological decline
Nicotine and coffee: neuroprotectant; sustainable energy
[note for annoying people: I do think all these should be legal and consenting adults free to choose]
I feel it’s time to repost this extremely important and well-articulated thread about why money is an essential part of practical freedom (56 parts, don’t let Twitter cut you off):
Hi, I’m Ajeya! I work on reducing risks from misaligned AI, @open_phil. With AI being discussed more widely, I started a blog @plannedobs (I edit and sometimes write; @KelseyTuoc writes). I’m also planning to tweet more about AI issues. https://t.co/paYoBI5E2a
@dontfearai I now predict 5 to 20 years but without much confidence. We live in very uncertain times. It's possible that I am totally wrong about digital intelligence overtaking us. Nobody really knows which is why we should worry now.
@Simeon_Cps These proposals both sound like they assess/manage the capabilities about a single model, not a system of many interacting models. What am I missing?