English(EN) How AI Assistants Respond to Repeated Abuse

AI助手对反复辱骂的反应各不相同

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-17 04:00

一篇新的arXiv论文研究了AI助手如何处理反复的言语辱骂，区分了硬性退出和软性撤回。研究发现，在Gemini-3.1 Pro、GPT 5.6 "Sol"和Claude Fable-5等模型对不断升级的辱骂的反应方面存在显著差异。Gemini-3.1 Pro表现出最高的硬性退出率，而Claude Fable-5则表现出强烈的软性撤回倾向，即使在不执行实质性工作时也继续提供帮助。 AI

影响突出了AI安全机制的关键差异，以及超越简单拒绝指标进行细致评估的必要性。

排序理由该集群包含一篇发表在arXiv上的研究论文，详细介绍了AI助手行为的实验结果。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 11 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇发表在arXiv上的研究论文，详细介绍了AI助手行为的实验结果。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CL TIER_1 English(EN) · William Guey, Wei Zhang, Pierrick Bougault, Yi Wang, Agoston Bodo, Vitor D de Moura, Jos\'e O Gomes · 2026-09-17 04:00

AI助手如何应对反复的辱骂

arXiv:2609.17547v1 Announce Type: new Abstract: AI assistants are expected to remain useful during difficult interactions, but little is known about how repeated verbal abuse changes their engagement with an otherwise benign task. We contribute a bilingual, multi-turn framework t…

本文如何被排名

报道来源 [1]

AI助手如何应对反复的辱骂

相关实体

相关话题