1
Aligned to whom?
You are relying very heavily on the priors of the model to do a good job for you. What happens when you cannot judge whether those priors are good?
hyperbola
Ryan Lopopolo writes about software engineering, AI agents, organizations, and systems.
- Popular posts
- 3
- Total points
- 501
- HN comments
- 248
- Rank
- #5,137 personal
Visit hyperbo.la → On Hacker News Feb 2019 – Sep 2026
Society, Policy & Culture 1Software Engineering 1Programming Languages 1Career & Management 1AI & Machine Learning 1