Choose an institution to see its records and dossier, or choose a topic.
Through year 2027 · all records, including scheduled dates 2026 2025 2024 2023 2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010 2008 2005 1965 Show Organisations and infrastructure All types Organisation Person Document Concept Infrastructure Public body Reset filters
94 people, organisations and other records · 122 relationships · 34 claims about a valueEverything recorded up to the chosen year, whether or not still in effect · Undated records included
Claim and evidence Close ×
The post “Shah and Yudkowsky on alignment failures” (2022-02-28) contests Y09. In a chat with Rohin Shah, he argues that predicting human text about a domain well enough requires the model to learn internally a process that does the work of that domain. Copy claim link
Copy claim reference
Machine-checked, not reviewed by an editor Stated in source Supported
Supporting evidence: primary · 1 source
Date 2022-02-28 · publication Role Contesting, Earliest located Register (kind of text) Nonfiction Engagement Engages a premise. Under the message-level rule, the passage, read with the paragraph, message or dialogue turn it belongs to, gives a reason against the premise that next-token prediction is a shallow task that cannot require deep cognition. Scope and limits Premise addressed: Next-token prediction is a shallow task that cannot require deep cognition. The quotation is a turn attributed to Yudkowsky in a dialogue. Chat log from the Late 2021 MIRI Conversations (section dated November 13, year implied 2021), published on LessWrong 2022-02-28. Only the labelled Yudkowsky turn is quoted. Conditional: if a model achieved AGI-complete prediction of text about science, it would need internal machinery that does science. Not a claim about current models. Lightly edited transcript of a conversation held before the posting date; the date shown is the publication date.
The evidence 01 primary source Supports the claim
Shah and Yudkowsky on alignment failures ↗ LessWrong · 2022-02-28
Copy source reference
Cite the original source Cite the original document, not Selekoa. Check these details against it.
Copy APA
Download BibTeX Download RIS Download CSL-JSON
Source excerpt “what I'm saying is that for GPT-5 to successfully do AGI-complete prediction of human text about Science, gradient descent must identify some formless thing that does Science internally in order to optimize the outer loss function for predicting human text about Science” Location in source §23 'November 13 conversation', turn [Yudkowsky][15:06]
Machine-checked · 2026-10-06
Source details and check notes Authors Rohin Shah; Eliezer Yudkowsky Version Post body retrieved through the LessWrong GraphQL API on 2026-10-06 (SHA-256 92623cf38321439650742c0457ceed8a8d904430986703db0eaae240d87beba4), content version 1.1.0; quotations also compared with the posting-day version 1.0.0. Canonical address Open the canonical source ↗ Accessed 2026-10-06 Language English (en) Attribution Rohin Shah and Eliezer Yudkowsky, “Shah and Yudkowsky on alignment failures”, LessWrong, 2022-02-28. Reuse No open redistribution licence asserted. Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution. Check notes Candidate pack passage T09-a. The quotation (42 words) matched the retrieved expression (SHA-256 92623cf38321439650742c0457ceed8a8d904430986703db0eaae240d87beba4) when checked on 2026-10-06. It also matched the posting-day version 1.0.0. Pack version note: edited post, current version 1.1.0; posting-day check pending. Status and meaning Editorial status. Not reviewed by an editor . An exploratory record that no editor has accepted. This status does not say whether a person has read any linked passage. Each evidence record above shows how it was checked.
Relation. A passage of the document disputes the proposition or a premise it depends on. The record documents the text’s position, not who is right.
Date. LessWrong postedAt.
Record ID yk-y09-lw-2022-shah-and-yudkowsky-on-alignment-failures-c