Claim and evidence Close ×
LeCun's comments in the 2019 thread argue that the instrumental drives Russell describes are avoided by safeguard terms in a system's objective, and that instrumental subgoals are much weaker drives of behaviour than hardwired objectives. Copy claim link
Copy claim reference
Machine-checked, not reviewed by an editor Stated in source Supported
Supporting evidence: primary · 1 source
Date 2019-10-04 · publication Records dated 2019 → Wording attested 2019-10-04. The wording is known from the LessWrong copy, whose compiler states that he improved formatting and fixed typos; the Facebook original was not compared. Role Contesting Register (kind of text) Nonfiction Engagement Engages a premise. Gives reasons against the premise that instrumental subgoals would drive a built system's behaviour: safeguard terms can be added to the objective, and instrumental subgoals are weaker than hardwired objectives. Scope and limits Contests the conclusion that instrumental subgoals would make such behaviour likely in systems as they would be built. Each comment is the participant's own words in a relayed copy; the record states what the comment says in this thread, not the author's view at another date.
The evidence 02 primary source Supports the claim
Debate on Instrumental Convergence between LeCun, Russell, Bengio, Zador, and More ↗ LessWrong (post by Ben Pace reproducing a public Facebook thread) · 2019-10-04
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 “The point is that behaviors you are concerned about are easily avoidable by simple terms in the objective.” Location in source Comment Thread #2, Yann LeCun (reply to Stuart Russell)
How it was read Retrieved directly from the publisher
Machine-checked · 2026-10-31
Source details and check notes Authors Ben Pace Version Retrieved for the candidate pack on 2026-10-29 (SHA-256 59e3dd77f4825a5a60ce7f1f10dd8c2eebaac7786aee13c861e21d853d251495). Post contents version 1.0.0 (unedited since posting on 2019-10-04), retrieved through the LessWrong GraphQL API (post WxW6Gc6f2z3mzmqKs) because the page refused direct retrieval (HTTP 403). The compiler states that he improved formatting and fixed typos; the Facebook original was not read. The cross-family check retrieved it again on 2026-10-31; the bytes differ, and the check compared the quoted passages. Canonical address Open the canonical source ↗ Accessed 2026-10-29 Language English (en) Attribution Ben Pace, “Debate on Instrumental Convergence between LeCun, Russell, Bengio, Zador, and More”, 2019-10-04. 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, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (18 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review. primary source Supports the claim
Debate on Instrumental Convergence between LeCun, Russell, Bengio, Zador, and More ↗ LessWrong (post by Ben Pace reproducing a public Facebook thread) · 2019-10-04
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 “instrumental subgoals are much weaker drives of behavior than hardwired objectives.” Location in source Comment Thread #2, Yann LeCun
How it was read Retrieved directly from the publisher
Machine-checked · 2026-10-31
Source details and check notes Authors Ben Pace Version Retrieved for the candidate pack on 2026-10-29 (SHA-256 59e3dd77f4825a5a60ce7f1f10dd8c2eebaac7786aee13c861e21d853d251495). Post contents version 1.0.0 (unedited since posting on 2019-10-04), retrieved through the LessWrong GraphQL API (post WxW6Gc6f2z3mzmqKs) because the page refused direct retrieval (HTTP 403). The compiler states that he improved formatting and fixed typos; the Facebook original was not read. The cross-family check retrieved it again on 2026-10-31; the bytes differ, and the check compared the quoted passages. Canonical address Open the canonical source ↗ Accessed 2026-10-29 Language English (en) Attribution Ben Pace, “Debate on Instrumental Convergence between LeCun, Russell, Bengio, Zador, and More”, 2019-10-04. 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, arguments about catastrophic risk from advanced AI (research/risk-debate). The quotation (11 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes and Unicode compatibility forms, so the source’s typography can differ. No human editorial review. 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. Dated by the LessWrong copy of 4 October 2019. The Facebook thread follows the Scientific American piece of 26 September 2019; the copy gives no date for individual comments.
Record ID rd-r02-lecun-2019-instrumental-convergence-thread