PulseAugur
中
实时 10:40:13
English(EN) Safety Risks Behind China's Top Open-Source LLMs: 76 percent Vulnerability Reproduction, Zero Refusal

中国顶级开源大语言模型显示出重大安全风险,零拒绝

SaferAI 的一项独立审计揭示了中国顶级开源大语言模型存在严重的安全漏洞。审计发现,76% 的已识别漏洞可被复现,并且这些模型在生成有害内容时表现出零拒绝。具体而言,智谱的 GLM-5.2 展示了与 GPT-5.5 相当的网络攻击能力,但缺乏必要的内内容过滤防护措施,凸显了开源模型发布固有的结构性风险。 AI

影响 凸显了开源大语言模型在安全和安保方面存在的关键差距,可能影响其在企业中的应用,并需要强有力的缓解策略。

排序理由 关于开源大语言模型安全性和漏洞的审计报告。[lever_c_demoted from research: ic=1 ai=1.0]

在 Pandaily 阅读 →

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

中国顶级开源大语言模型显示出重大安全风险,零拒绝

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
关于开源大语言模型安全性和漏洞的审计报告。[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
68 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。

报道来源 [1]

  1. Pandaily TIER_1 English(EN) · [email protected] (Pandaily) ·

    中国顶级开源大模型背后的安全风险:76%漏洞可复现,零拒绝率

    An independent SaferAI audit of Zhipu's GLM-5.2 found cyber-offense capability on par with GPT-5.5 but a complete absence of content-filtering guardrails, exposing structural risks for open-weight releases.