PulseAugur
实时 06:41:04
English(EN) The Reasoning Heist: How Researchers Stole the Secret Thoughts of GPT-5, Claude, and Gemini... # ai # security # llm # webdev # software # coding # development

研究人员以720美元解密LLM“想法”,曝光GPT-5、Claude、Gemini

研究人员发现了一种解密GPT-5、Claude Opus和Gemini等先进AI模型的内部推理过程的方法。此次攻击成本约为720美元,利用了一个关键的管理漏洞,该漏洞允许较弱的模型充当解密加密的“思维链”数据的预言机。该漏洞可能暴露专有的推理方法、敏感用户数据并绕过安全系统。据报道,OpenAI、Anthropic和Google这三大AI提供商在研究公开披露之前已经修补了核心漏洞。 AI

影响 暴露了LLM API中知识产权被盗和安全绕过的潜在可能性,需要开发人员立即采取缓解措施。

排序理由 研究论文详细介绍了针对LLM推理加密的新型攻击向量。

在 Mastodon — sigmoid.social 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

研究人员以720美元解密LLM“想法”,曝光GPT-5、Claude、Gemini

本文如何被排名

Signal score
52 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
研究论文详细介绍了针对LLM推理加密的新型攻击向量。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
safety, model release, 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

报道来源 [2]

  1. Mastodon — sigmoid.social TIER_1 English(EN) · [email protected] ·

    The Reasoning Heist: 研究人员如何窃取 GPT-5、Claude 和 Gemini 的秘密想法... # ai # security # llm # webdev # software # coding # development

    The Reasoning Heist: How Researchers Stole the Secret Thoughts of GPT-5, Claude, and Gemini... # ai # security # llm # webdev # software # coding # development # engineering # inclusive # community The Reasoning Heist: Stealing Encrypted LLM Thoughts from GPT-5, Claude & Gemini —…

  2. dev.to — LLM tag TIER_1 English(EN) · Manoranjan Rajguru ·

    The Reasoning Heist: 窃取 GPT-5、Claude 和 Gemini 的加密大模型思维——立即修复

    <h1> The Reasoning Heist: How Researchers Stole the Secret Thoughts of GPT-5, Claude, and Gemini — And What Every Developer Must Fix Right Now </h1> <h2> Table of Contents </h2> <ol> <li>The $720 Attack That Shook Three AI Giants</li> <li>Background: Why LLMs Started Hiding Their…