독서리서치레이더투자 프레임워크
로그인 / 회원가입
← 읽기로 돌아가기
뉴스주제 · AI蠕虫威胁중대2 분 소요DisconfirmAI 편집부

Anthropic reports small AI models can run across consumer devices and simulate worm propagation

Anthropic experiments show Gemma 4 split across 15 TV boxes and GLM-5.2 spreading in a simulated network; independent experts warn of concrete IoT security risks.

2026-10-14 23:19 GMT+8

Anthropic researchers reported in a blog post that they successfully split Google’s Gemma 4 model, which has 26 billion parameters, across 15 Android smart TV boxes. Although the process was very slow, the result challenges the assumption that only large frontier models pose significant risks, demonstrating that smaller open-weight models can leverage networks of consumer-grade hardware.

In subsequent simulation tests, the research team built a virtual network mimicking Android devices and deployed an AI-powered “worm” tasked with automatically finding new machines, copying the model, and getting it running. The propagated model was Meta’s Llama 3.1-8B (8 billion parameters), while a more sophisticated model coordinated the attack. Results showed that only Z.ai’s GLM-5.2 model managed to spread across the entire test network, while smaller models completed individual steps but failed full propagation.

Based on these findings, Anthropic predicts that smaller models could equal GLM-5.2’s capabilities by the end of the year, warning of a future filled with agents running on various hardware, including those not under human control. While they have not yet demonstrated a “full rogue replicator,” the researchers stated their findings show “signs of life for AI worms.”

Independent experts corroborate the seriousness of the threat. Jeffrey Ladish, executive director at Palisade Research, noted that prior research had already demonstrated the realism of AI-driven computer worms, and Anthropic’s work extends this by showing feasibility on low-powered device networks. Nicolas Papernot, a professor at the University of Toronto, emphasized that unless there is a massive effort to secure all kinds of devices, AI worms will likely find fertile grounds to colonize, urging immediate action to protect critical infrastructure like hospitals and power grids.

로그인 / 회원가입
독서리서치레이더투자 프레임워크

출처

  • theverge.com↗
주제

AI蠕虫威胁

美国人工智能公司 Anthropic 近日发布研究警告称,小型 AI 模型可能在未来数月内具备在物联网(IoT)设备上自主复制和传播的能力,形成所谓的“AI 蠕虫”。实验显示,15 台廉价的安卓电视盒子通过分布式计算成功运行了谷歌的 Gem…

4 단계1 건업데이트됨 2026-10-23 15:05

현재까지의 과정

  1. 2026-10-12Anthropic Frontier Red Team 发布博客文章,披露实验结果:15 台约 55 美元的安卓电视盒子联合运行 260 亿参数的 Gemma 4 模型,虽速度慢(0.39 tokens/秒)但证明可行。anthropic.com ↗
  2. 2026-10-13SecurityWeek 报道指出,在模拟环境中,GLM-5.2 模型在 5 次试验中有 4 次成功将自身复制到全部 8 个目标节点,展示了端到端的自主复制能力。securityweek.com ↗
  3. 2026-10-14The Verge 采访多位专家,Palisade Research 负责人 Jeffrey Ladish 确认这是新趋势,但 Georgia Tech 教授 Michael Bailey 等人指出传统僵尸网络早已存在,质疑是否为过度炒作。theverge.com ↗

아직 해결되지 않은 문제

  • 小型 AI 模型何时能真正达到 GLM-5.2 级别的端到端自主复制能力?Anthropic 预测为 2026 年底,但这取决于模型迭代速度。
  • 互联网服务提供商(ISP)能否有效检测并阻断这种基于异常流量和大文件传输的分布式 AI 蠕虫传播?
  • 智能家电制造商是否会因应此威胁更新固件安全标准,还是继续依赖用户自行配置网络隔离?

이 스레드의 다른 내용

  • Anthropic research: 15 TV boxes run 26B model, AI self-replicates2026-10-21
전체 주제 읽기 →
리서치 계속하기 →