"theory uplift differentially benefits safety & is massively underpriced" by Yudhister Kumar
About this episode
Yudhister Kumar discusses the rapid development of AI mathematics capabilities and its implications for safety and understanding model behavior.
[1] We will likely have near-superhuman mathematics AI by Q1 2027. [1] [2] Qualitatively, AI mathematics capabilities are developing significantly faster than automated AI R&D capabilities. [2] [3] Thus, we will likely have a period of time where the rate of our ability to rigorously & usefully verify and understand model behavior and model outputs outpaces the rate of capability development itself. [4] Our ability to take advantage of this period is bottlenecked on the qu...
More episodes of LessWrong (Curated & Popular)
- "Lighthaven East - A Feasibility Study" by JohnofCharleston
- "Empowerment, corrigibility, etc. are simple abstractions (of a messed-up ontology)" by Steven Byrnes
- "Trees are mostly made of air and a generalizable lesson for AI safety" by zroe1
- "Mnemonic portraits for 19,023 human genes" by Brinedew
- "Cognitive Security as an AI Safety Cause Area" by jsteinhardt
- "Women should be able to open things" by KatjaGrace
Explore listener stats, chart rankings, contacts and more on the LessWrong (Curated & Popular) podcast page.