A breakdown of AI capability levels focused on AI R&D labor acceleration — LessWrong
lesswrong.com · 4,270 words · saved by 1 readers
In a variety of conversations about AI misalignment risks, I find that it is important to be able to clearly point at different levels of AI capabili…
x A breakdown of AI capability levels focused on AI R&D labor acceleration — LessWrong AI Frontpage 122 A breakdown of AI capability levels focused on AI R&D labor acceleration by ryan_greenblatt 22nd Dec 2024 AI Alignment Forum 7 min read 16 122 Ω 57 In a variety of conversations about AI misalignment risks, I find that it is important to be able to clearly point at different levels of AI capability. My current favorite approach is to talk about how much the AI accelerates AI R&D [1] labor. I define acceleration of AI R&D labor by Y times as "the level of acceleration which is as useful (for
related reading
- My picture of the present in AI — LessWronglesswrong.com
- Current AIs seem pretty misaligned to me — LessWronglesswrong.com
- Will AI R&D Automation Cause a Software Intelligence Explosion?forethought.org
- Differential acceleration of alignment-relevant capabilities is a bad bet — LessWronglesswrong.com
- Does AI Progress Have a Speed Limit?—Asteriskasteriskmag.com
- My picture of the present in AI - by Ryan Greenblattblog.redwoodresearch.org
- How do we (more) safely defer to AIs? — LessWronglesswrong.com
- The least understood driver of AI progress | Epoch AIepoch.ai
- I Would Have Solved Alignment, But I Was Worried That Would Advance Timelines — LessWronglesswrong.com
- Plans A, B, C, and D for misalignment risk — LessWronglesswrong.com
- AI safety undervalues founders — LessWronglesswrong.com
- Defining alignment research — LessWronglesswrong.com