Incoming Berkeley professor lists open questions: pre/post-training alignment, safe agents
canondetortugas · x · 2026-10-15
An incoming UC Berkeley assistant professor lists his open research questions: what mechanisms let pre-training produce models good enough for post-training to succeed, how to train verifiably safe-by-design agents, what makes data learnable (predicting scaling laws from intrinsic complexity), and how all this extends beyond language.
More from AGI Musings
- Arguments denying AI intelligence would also deny many animals and people, writer argues — NathanpmYoung · 2026-10-15
- Matt Hartman, who wrote Hugging Face's first check, explains his play-first investing method — danshipper · 2026-10-15
- "You distilled humanity": devs slam Anthropic's double standard on training data and model output — HankYeomans · 2026-10-15
- Blaise Aguera: Human intelligence is societal—and top AI works as societies of models — blaiseaguera · 2026-10-15
- Security researcher: doomers don't belong in security — HankYeomans · 2026-10-15
- New Paper: 500k Glassdoor Reports Show Interviews Are Two-Way Screening — Bad News for AI Interviews — danielrock · 2026-10-15