Researcher tells AI labs: theory of DL is the highest-ROI bet and the blocker to RSI

jasondeanlee · x · 2026-11-21

jasondeanlee argues that the highest-ROI work for AI labs is developing a theory of deep learning and AI, calling it the blocker to recursive self-improvement—solve it and labs would never need to hire a new AI researcher again, which he calls "the Anthropic dream." He jokes that his students are job-hunting, and so is he, "but I need $100M."

Original post →

More from AGI Musings

AGI Musings channel →