UC Berkeley RDI(@BerkeleyRDI):We have seen plenty of self-improving AI. We have not seen recursive self-improvement. @OriolVinyalsML, Co-founder & CTO of Discovery Loop and former VP of Research at Google DeepMind, draws the line at whether the system can rewrite its own harness and its own weights, not just its output. Has anything crossed that line yet? If not, how do we get there? Drop your take below. 👇

UC Berkeley RDI

@BerkeleyRDI

UC Berkeley's campus-wide, cross-disciplinary Center for Responsible, Decentralized Intelligence - RDI

Joined December 2021

48 Following 4.3K Followers

UC Berkeley RDI@BerkeleyRDI

2026.09.24 16:04

We have seen plenty of self-improving AI. We have not seen recursive self-improvement. @OriolVinyalsML, Co-founder & CTO of Discovery Loop and former VP of Research at Google DeepMind, draws the line at whether the system can rewrite its own harness and its own weights, not just its output. Has anything crossed that line yet? If not, how do we get there? Drop your take below. 👇