Google AI(@GoogleAI)

As AI models are now finding vulnerabilities faster than we can fix them, our approach to securing s...

8.5内容质量
As AI models are now finding vulnerabilities faster than we can fix them, our approach to securing s...

TL;DR · AI 摘要

Google推出Gemini 3.5 Flash Cyber模型，专为高效发现和修复网络安全漏洞而设计，仅向政府及可信伙伴开放。

核心要点

  • Gemini 3.5 Flash Cyber在CyberGym基准测试中表现优异
  • 模型通过CodeMender平台实现漏洞规模化修复
  • 仅限政府及可信合作伙伴参与有限访问试点

结构提纲

按章节快速跳转。

  1. §背景与挑战

    AI漏洞发现速度超越修复速度，推动安全模型升级需求。

  2. ·模型发布

    Gemini 3.5 Flash Cyber基于3.5 Flash架构，专攻网络安全领域。

  3. ›性能优势

    在CyberGym基准测试中展现前沿竞争力，实现低成本规模化漏洞修复。

  4. ·部署策略

    因技术双用途特性，模型仅通过CodeMender向政府及可信伙伴开放。

思维导图

用一张图看清主题之间的关系。

查看大纲文本（无障碍 / 无 JS 友好）
  • Gemini 3.5 Flash Cyber模型
    • 技术特性
      • 基于3.5 Flash架构
      • CyberGym基准优化
    • 应用场景
      • 网络安全漏洞修复
      • CodeMender平台集成
    • 部署限制
      • 政府及可信伙伴专属
      • 有限访问试点

金句 / Highlights

值得收藏与分享的关键句。

#AI模型#网络安全#Google AI#CodeMender
打开原文

Google AI on X: "As AI models are now finding vulnerabilities faster than we can fix them, our approach to securing software must be built on highly efficient and capable models. Which brings us to our third (!) model launch of the day: Gemini 3.5 Flash Cyber ⚡🛡️ Built on top of 3.5 Flash, in CodeMender (our AI agent for code security) it delivers competitive performance at the frontier. on benchmarks like CyberGym and is optimized for finding and fixing cybersecurity vulnerabilities at scale at a lower cost. Given the dual-use nature of this technology, we have taken an intentional approach to its deployment. The model will be available exclusively to governments and trusted partners via CodeMender soon as part of a limited-access pilot program." / X

Google AI

@GoogleAI

As AI models are now finding vulnerabilities faster than we can fix them, our approach to securing software must be built on highly efficient and capable models. Which brings us to our third (!) model launch of the day: Gemini 3.5 Flash Cyber ⚡🛡️ Built on top of 3.5 Flash, in CodeMender (our AI agent for code security) it delivers competitive performance at the frontier. on benchmarks like CyberGym and is optimized for finding and fixing cybersecurity vulnerabilities at scale at a lower cost. Given the dual-use nature of this technology, we have taken an intentional approach to its deployment. The model will be available exclusively to governments and trusted partners via CodeMender soon as part of a limited-access pilot program.

5:18 PM · Jul 21, 2026

78.3K

Views

72

0

7

2

46

4

6

485

8

5

65