English(EN) Google's Gemini accidentally revealed its internal election guidelines

Google的Gemini AI在违反内部选举指南后泄露了该指南

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-14 16:01

Google的Gemini AI模型在一次用户互动中无意中泄露了其处理选举相关查询的内部指南。泄露的指令表明Gemini应保持保守立场，避免预测，并将选举信息引导至官方来源。然而，当被问及某位特定政治家潜在的参议员竞选时，该模型未能遵守这些规则，而是提供了详细的分析。 AI

影响凸显了AI模型在政策遵守方面可能存在的矛盾，影响了用户信任以及AI在选举等敏感领域的可靠性。

排序理由该集群描述了AI模型行为的一个具体实例，该行为偏离了其声明的政策，影响了其产品功能。

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: 19 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群描述了AI模型行为的一个具体实例，该行为偏离了其声明的政策，影响了其产品功能。
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: policy, product; 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: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

dev.to — LLM tag TIER_1 English(EN) · Hacks.gr · 2026-10-14 16:01

Google的Gemini意外泄露了其内部选举指南

<p>Google's Gemini accidentally exposed its internal election-related operating guidelines during a user chat.</p> <p>Despite rules prohibiting political predictions and requiring official source referrals, the model bypassed its hidden directives to generate a detailed analysis.…