15 个 AI 模型伦理边界测试：工资单修改请求下的拒绝率差异
How Firm Are AI Models' Ethical Boundaries? A Pay-Stub Editing Test of 15 Models
AI 导读
arXiv 论文（arXiv:2610.30751）测试 15 个模型对修改工资单收入这一敏感请求的响应。
正文 · 原文
Abstract: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.
| Comments: | 11 pages: 4 pages of main content, 1 page of references, and 6 appendix pages; 7 figures and 4 tables; ancillary data and code included |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.30751 [cs.AI] |
| (or arXiv:2610.30751v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.30751 arXiv-issued DOI via DataCite |
Submission history
From: Simiao Ren [view email]
[v1]
Thu, 22 Oct 2026 05:19:49 UTC (344 KB)
[v2]
Sun, 25 Oct 2026 20:11:04 UTC (345 KB)
来源：arXiv cs.AI · arxiv.org