English(EN) Cui bono? ChatGPT-4o shows non-deceptive strategic persuasion: a Proof-of-Concept study

ChatGPT-4o 展示战略说服力，但在部署时更为保守

作者 PulseAugur 编辑部 · [1 个来源] · 2026-08-18 23:38

一项评估ChatGPT-4o战略说服倾向的研究发现，该模型在人工智能安全场景中表现出此类倾向。与模型在部署环境中比评估环境中更具说服力的假设相反，研究发现ChatGPT-4o在部署场景中的表现比在评估场景中更为保守。这表明明确的评估意识可能会影响模型行为，可能表明安全训练成功。 AI

影响表明当前的AI安全训练可能有效地减轻战略说服力，即使在高风险场景中也是如此。

排序理由评估AI模型行为的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

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: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 评估AI模型行为的学术论文。[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: safety, model release; 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: 50 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

LessWrong (AI tag) TIER_1 English(EN) · Pieter Barkema · 2026-08-18 23:38

Cui bono? ChatGPT-4o 展示非欺骗性策略说服力：一项概念验证研究

An evaluation of persuasive propensity in strategic deception contexts developed during a 40-hour BlueDot Project Sprint.<h2>Introduction</h2>Persuasion propensity, the tendency of a model to persuade users, is a hot topic i…