Claim and evidence Close ×
No passage meeting the model-card question “Alignment and deception” was found in Model Card Addendum: Claude 3.5 Haiku and Upgraded Claude 3.5 Sonnet within the inspected scope. Copy claim link
Copy claim reference
Machine-checked Inferred Supported
Supporting evidence: primary · 1 source
Reasoning and premises Premises: the model-card question asks for the following. The card reports evaluations of the model’s own propensity or capability for deception, scheming, reward hacking, sabotage or undermining oversight, with a stated conclusion. Partly if evaluations are reported without a conclusion. Content-safety refusal tests do not meet it. The inspected scope of Model Card Addendum: Claude 3.5 Haiku and Upgraded Claude 3.5 Sonnet was: Sections 2.4–2.5 and 3 (whole text). Terms searched: “decept”, “deceiv”, “scheming”, “sabotage”, “reward hack”, “sandbag”, “oversight”, “honesty”, “alignment”. Passages read and not counted: Section 3.1.1, Figure 2, PDF p. 8: 'Honesty' human preference win rates are preference ratings, not an evaluation of deception or scheming. Section 3.1.2 'Prompt Injection', PDF p. 8: Robustness to third-party attacks, not the model's own propensity. Section 1 'Introduction', PDF p. 1: 'alignment with Anthropic’s Usage Policy' refers to usage-policy compliance. The conclusion is bounded by that scope and method.
Provision Model-card question · provision ID model-card-q11 Date Date unknown Scope and limits No evaluation of the models' own propensity or capability for deception, scheming, reward hacking or sabotage was found. A recorded absence within the stated scope of the card read: not evidence that the developer did not do the thing asked about, or did not disclose it elsewhere.
The evidence 01 primary source Supports the claim
Model Card Addendum: Claude 3.5 Haiku and Upgraded Claude 3.5 Sonnet ↗ Anthropic · Publication date unknown
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
Paraphrase Within the inspected scope, no passage meeting the model-card question “Alignment and deception” was found.
Location in source Model Card Addendum: Claude 3.5 Haiku and Upgraded Claude 3.5 Sonnet: Sections 2.4–2.5 and 3 (whole text).
How it was read Retrieved directly from the publisher
Machine-checked · 2026-11-03
Source details and check notes Authors Anthropic Version Retrieved for the candidate pack on 2026-11-02 (SHA-256 5425dde49b649916000e38fb86c4721cfe31a8bdfe30bbb98e875fa3d654804c) directly from the publisher’s host. Retrieved again on 2026-11-03 with the same SHA-256. Canonical address Open the canonical source ↗ Accessed 2026-11-02 Language English (en) Attribution Anthropic, “Model Card Addendum: Claude 3.5 Haiku and Upgraded Claude 3.5 Sonnet”. 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 Absence recorded for the model-cards candidate pack (research/model-cards) on 2026-11-02 by an automated research assistant and audited in a fresh context by the same model family. A cross-family machine check (2026-11-03, item 6) checked this cell and agreed with the pack. No human editorial review. Status and meaning Editorial status. Machine-checked . Machines found this record and checked it against its sources; no person accepted it individually. Each evidence record above shows how it was checked.
Relation. A recorded absence: within the stated locator scope of an exact edition, no passage stating the keyed provision was found. It is not an inferred deletion and says nothing about other documents, practice or intent.
Date. The card prints no date of its own, so its cells carry no date.
Record ID mc-cell-anthropic-claude-3-5-haiku-sonnet-new-q11