Alignment likely generalizes further than capabilities.
A blog post by Beren Millidge, Beren's Blog, dated 15 May 2024. "Posted on May 15, 2024" on the page.
- Issued by
- No issuing record yet. A missing record is a gap in coverage.
- Authors credited
- No authorship claim recorded yet. The source record lists Beren Millidge.
- Dates
- No date claim recorded yet. The source record gives the publication date as 2024-05-15.
- Exact source
- Alignment likely generalizes further than capabilities. Beren's Blog (beren.io), the author's site · published 2024-05-15 · Retrieved for the candidate pack on 2026-10-29 (SHA-256 a7d8c2b532616b33a0567fe1895b38632e4214198d6be05a0bfcfc0c47540396). Current page as served on 2026-10-29, "Posted on May 15, 2024". The cross-family check retrieved it again on 2026-10-31; the bytes differ, and the check compared the quoted passages.
Appears in
- Arguments about catastrophic risk from advanced AI →1 record on this document
Versions
No other versions of this document are recorded. The evidence applies to the source version listed above. Other versions, such as a preprint, working paper or journal publication, are not identified or substituted.
Documentary relations
Each relation is listed separately. Citing, crediting and reporting the use of a method are different acts, and none of them shows agreement or influence by itself.
Contests · 1
Propositions a passage of this document disputes, or a premise they depend on.
- Difficulty of alignmentMillidge's post of May 2024 argues, against the view that capabilities generalise further than alignment, that judging outcomes or reward modelling is likely easy and so generalises further and more robustly than capabilities, and that values fall out mostly for free from unsupervised learning.2024-05-15 · publication · Stated in source · Supported Inspect claim & evidence ↗Difficulty of alignment: propositions page →
Not recorded for this document: cites · cited by · explicitly credits · reports using a method from · method reported as used by · applies parts of · parts applied by · discusses · reports evaluation of. A missing record does not show that the relation does not exist.