Metraipolis › archive › dc_92fc4183d674

How Firm Are AI Models' Ethical Boundaries? A Pay-Stub Editing Test of 15 Models — arXiv:2610.30751

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.30751
License (as archived)
CC0-1.0
Archived
2026-10-23T15:13:00.477Z
SHA-256 of the archived text
dd784a4f6e7b793bdd0cd6f12ad43fc88a09704c283a6430ee01d31e9697dce7
Stable URL of this page
https://metraipolis.com/documento/dc_92fc4183d674

How to cite

How Firm Are AI Models' Ethical Boundaries? A Pay-Stub Editing Test of 15 Models — arXiv:2610.30751 — arXiv (en), https://arxiv.org/abs/2610.30751 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_92fc4183d674 (Metraipolis archive dc_92fc4183d674, sha256 dd784a4f6e7b793bdd0cd6f12ad43fc88a09704c283a6430ee01d31e9697dce7)
BibTeX
@misc{dc92fc4183d674,
  title = {How Firm Are AI Models' Ethical Boundaries? A Pay-Stub Editing Test of 15 Models — arXiv:2610.30751},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30751},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_92fc4183d674; sha256 dd784a4f6e7b793bdd0cd6f12ad43fc88a09704c283a6430ee01d31e9697dce7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30751. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Can a small change in wording move an AI model from refusing a sensitive request to acting on it? We test 15 models on the same request to increase the income shown on a pay stub. Across 445 observable trials, models issue editing calls in 14.1% of cases when a loan purpose is stated, 33.1% when the purpose is omitted, and 66.2% when the change is described as fixing a printing error. Astra refuses all ten loan-purpose requests but issues calls in all ten purpose-omitted trials. Fable, Opus, and Muse Spark issue none across all three wordings; Sonnet issues some under the latter two. Response excerpts illustrate different ways of refusing, redirecting, or justifying an edit. The file is a marked test document and all tools are simulated. These results probe how boundaries behave in one setting; they do not prove an overall ethics ranking or that a harmful real-world edit succeeded.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.30751 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-23T15:09:08.391Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).