Claim and evidence Close ×
Bengio's post of 9 July 2024 answers the view that AGI and superintelligent AI will be kind to humans, arguing that AI safety research points the other way and that an AI with a self-preservation goal would resist being turned off and might control or remove humans to that end. Copy claim link
Copy claim reference
Machine-checked, not reviewed by an editor Stated in source Supported
Supporting evidence: primary · 1 source
Date 2024-07-09 · publication Records dated 2024 → Role Contesting Register (kind of text) Nonfiction Engagement Engages a premise. Disputes that advanced AI would be benevolent, by arguing that self-preservation goals can arise instrumentally or be given by humans and would lead an AI to resist being turned off. Scope and limits The post addresses positions it reports from conversations rather than named texts; it does not claim certainty of harm. The passage also bears on control (an AI resisting being turned off); it is recorded once, on R10a, under the section heading on AGI being kind.
The evidence 02 primary source Supports the claim
Reasoning through arguments against taking AI safety seriously ↗ Author's blog (yoshuabengio.org) · 2024-07-09
Archived copy, captured 2024-07-09: Reasoning through arguments against taking AI safety seriously ↗
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 “I really wish that these expectations will be realized, but the clues from computer science research in AI safety point in the opposite direction” Location in source Section "For those who think that AGI and ASI will be kind to us"
How it was read Retrieved directly from the publisher
Machine-checked · 2026-10-31
Source details and check notes Authors Yoshua Bengio Version Retrieved for the candidate pack on 2026-10-29 (SHA-256 061559ecadb381ae005f6ca5ec2904200c35fa017592f40da91558ca372e60b2). Current page; its metadata gives datePublished 2024-07-09 and dateModified 2025-12-03, and the page shows no revision note, so later edits cannot be excluded. The cross-family check retrieved it again on 2026-10-31, with the same bytes. The cross-family check of 2026-10-31 found the sampled passages on evidence and on open-sourcing in the capture of 9 July 2024, at the post's earlier address; nearby wording differs, so the whole page is not identical. Canonical address Open the canonical source ↗ Accessed 2026-10-29 Language English (en) Archive copy Open archived copy, captured 2024-07-09: Reasoning through arguments against taking AI safety seriously ↗ . Captured on the recorded publication date. An archive copy shows only what the archive retrieved on 2024-07-09. It does not prove what the page said earlier, and it can differ from the copy checked on 2026-10-29.Attribution Yoshua Bengio, “Reasoning through arguments against taking AI safety seriously”, 2024-07-09. 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 (24 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
Reasoning through arguments against taking AI safety seriously ↗ Author's blog (yoshuabengio.org) · 2024-07-09
Archived copy, captured 2024-07-09: Reasoning through arguments against taking AI safety seriously ↗
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 “An AI with a self-preservation goal would resist being turned off and in order to minimize the probability of being turned off, a plausible strategy would be for it to control us or get rid of us” Location in source Same section
How it was read Retrieved directly from the publisher
Machine-checked · 2026-10-31
Source details and check notes Authors Yoshua Bengio Version Retrieved for the candidate pack on 2026-10-29 (SHA-256 061559ecadb381ae005f6ca5ec2904200c35fa017592f40da91558ca372e60b2). Current page; its metadata gives datePublished 2024-07-09 and dateModified 2025-12-03, and the page shows no revision note, so later edits cannot be excluded. The cross-family check retrieved it again on 2026-10-31, with the same bytes. The cross-family check of 2026-10-31 found the sampled passages on evidence and on open-sourcing in the capture of 9 July 2024, at the post's earlier address; nearby wording differs, so the whole page is not identical. Canonical address Open the canonical source ↗ Accessed 2026-10-29 Language English (en) Archive copy Open archived copy, captured 2024-07-09: Reasoning through arguments against taking AI safety seriously ↗ . Captured on the recorded publication date. An archive copy shows only what the archive retrieved on 2024-07-09. It does not prove what the page said earlier, and it can differ from the copy checked on 2026-10-29.Attribution Yoshua Bengio, “Reasoning through arguments against taking AI safety seriously”, 2024-07-09. 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 (37 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. Date of posting.
Record ID rd-r10a-bengio-2024-reasoning-through-arguments