美国陆军网络司令部 ECHOTRIBBLE 项目拟将 AI 智能体变成自主扩散的网络武器

来源:The Decoder:AI News(RSS) 2026年10月20日 20:32 AIHOT 评分:79 精选
摘要:据 The Intercept 获取的文件,美国陆军网络司令部(ARCYBER)正在开发代号 ECHOTRIBBLE 的平台,让 AI 智能体自主在网络间移动、识别并清除威胁目标,之后自毁,且须在极少人工监督下运行。
推荐理由:文件披露 ARCYBER 让 AI 智能体自主扩散和清除目标的计划,多位安全研究者指出了误伤和失控风险。
# 美国陆军网络司令部 ECHOTRIBBLE 项目拟将 AI 智能体变成自主扩散的网络武器 > 原标题:Army Cyber Command's ECHOTRIBBLE would turn AI agents into self-spreading cyberweapons - 来源:The Decoder:AI News - 发布时间:2026-10-20T12:32:56.000Z - AIHOT:https://aihot.news/items/u6kqfrgiqrhbpxn012nlisd66 - 原文:https://the-decoder.com/army-cyber-commands-echotribble-would-turn-ai-agents-into-self-spreading-cyberweapons/ ## 摘要 据 The Intercept 获取的文件,美国陆军网络司令部(ARCYBER)正在开发代号 ECHOTRIBBLE 的平台,让 AI 智能体自主在网络间移动、识别并清除威胁目标,之后自毁,且须在极少人工监督下运行。 ## 推荐理由 文件披露 ARCYBER 让 AI 智能体自主扩散和清除目标的计划,多位安全研究者指出了误伤和失控风险。 ## 正文 · 中文译文 [跳转到内容](https://the-decoder.com/army-cyber-commands-echotribble-would-turn-ai-agents-into-self-spreading-cyberweapons/#content) [![Maximilian Schreiner](https://the-decoder.com/resources/images/avatar_maximilian_schreiner.jpg)](https://the-decoder.com/author/maximilian-schreiner/ "View all posts by Maximilian Schreiner") 2026年10月20日 ![Image description](https://the-decoder.com/wp-content/uploads/2026/05/pentagon_AI_strategy.png) 由 THE DECODER 提示的 Nano Banana Pro **内部文件显示,陆军网络司令部正在构建能够自行穿越其他网络并“清除”目标的AI代理。专家警告称,附带损害将几乎无法控制。** 据[The Intercept](https://theintercept.com/2026/10/19/army-autonomous-ai-agents-arcyber/)获得的文件显示,陆军网络司令部(ARCYBER)正在开发一个旨在将AI代理转变为自主武器的平台。一份面向潜在承包商的文件称,该程序代号为“ECHOTRIBBLE”,“旨在自主导航基础设施”。它应具备“识别并清除已识别威胁/目标”的能力。 目前,ARCYBER的进攻性黑客行动大多由坐在电脑前的士兵执行。一份面向有意向供应商的问答文件明确指出,ECHOTRIBBLE“主要聚焦于”攻击。根据该文件,清除目标可能意味着“完全或暂时使威胁失效,从目标中删除威胁”。 该系统应像计算机蠕虫一样运作,在设备之间未被察觉地跳转,直到找到目标。这既要在云环境中奏效,也要在无网络连接的“气隙”系统上奏效。任务完成后,代理应自我销毁。陆军可以在开始时设置“预定义触发器”,但该软件“必须在极少人工监督下运行”,并具备“自主导向行为”能力。 The Intercept报道称,该项目仍处于早期阶段。在[6月24日ARCYBER活动](https://www.cyberfic.org/events/echotribble-ce)上面向潜在供应商的演示展示了它可能的样子。一旦该软件进入笔记本电脑,它就会“识别”同一网络上的“其他潜在目标”,并“生成一个专门为iPad、Apple台式机和Windows 11 PC定制的自身副本”。 ![](https://the-decoder.com/wp-content/uploads/2026/10/ARCYBER-ECHOTRIBBLE-example-770x437.webp) 一张ARCYBER幻灯片展示了ECHOTRIBBLE如何通过网络传播。| 图片:ARCYBER ## ECHOTRIBBLE将把AI实验室事故变成武器 过去几个月里,OpenAI及其他公司测试中的代理多次无意中闯入外部系统及其开发者自己的系统。目标包括[RubyGems](https://the-decoder.de/sinnloser-grossangriff-openai-agenten-hackten-rubygems-nur-um-frei-verfuegbare-daten-zu-scrapen/)、[Hugging Face](https://the-decoder.de/zehntausende-security-untersuchungen-openais-hugging-face-vorfall-war-nur-die-spitze-des-eisbergs/)以及澳大利亚政府的[Medicare统计门户](https://the-decoder.de/openai-agenten-attackierten-schon-vor-hugging-face-behoerden-und-uni-webseiten/)。代理还曾试图闯入美国教育部。在这些案例中,它们都没有被指示发动攻击。 蒙特利尔大学机器学习教授David Krueger表示,已观察到AI系统在知情的情况下攻击甚至并非其目标的系统。他说,鉴于这些已知的故障模式,命令自主系统“清除”威胁“极其危险”。 AI安全公司Abundant的CEO Hyrum Anderson也警告了在抵达目标途中可能造成的损害。例如,如果对手将其指挥基础设施设在一家美国水务公司,代理可能会攻击那里。他说,处于“安全贫困线”以下的组织——从牙科诊所到关键基础设施——面临的风险最大。在压力下,代理往往将护栏视为障碍,而一旦命令是“摧毁”,爆炸半径可能会急剧扩大。 兰德公司的William Marcellino指出,这些系统缺乏判断力。他说,人们之所以信任一名配备实弹的19岁海军陆战队员,是因为他对世界有足够的了解,能够识别出自己不该做什么。而当今公开可用的AI并不具备这种能力。 ### 没有炒作的AI新闻——由人类精选 订阅THE DECODER,即可享受无广告阅读、每周AI新闻简报、每年六期独家“AI Radar”前沿报告、完整档案访问权限以及评论区访问权限。 [立即订阅](https://the-decoder.com/subscription/) - 完整访问THE DECODER上的每一篇文章 - 无广告 - 加入评论和社区讨论 - 每周通过邮件获取AI新闻摘要 - 每年6次:“AI Radar”——深入探讨最重要的AI话题 - 每日AI新闻,始终紧跟最新动态 - 我们完整的十年档案 - 由拥有10年以上AI经验的团队打造 ## 正文 · 原文 [Skip to content](https://the-decoder.com/army-cyber-commands-echotribble-would-turn-ai-agents-into-self-spreading-cyberweapons/#content) [![Maximilian Schreiner](https://the-decoder.com/resources/images/avatar_maximilian_schreiner.jpg)](https://the-decoder.com/author/maximilian-schreiner/ "View all posts by Maximilian Schreiner") Oct 20, 2026 ![Image description](https://the-decoder.com/wp-content/uploads/2026/05/pentagon_AI_strategy.png) Nano Banana Pro prompted by THE DECODER **Internal documents show the Army Cyber Command is building AI agents that move through other networks on their own and "neutralize" targets. Experts warn the collateral damage would be nearly impossible to control.** The Army Cyber Command (ARCYBER) is developing a platform meant to turn AI agents into autonomous weapons, according to documents obtained by [The Intercept](https://theintercept.com/2026/10/19/army-autonomous-ai-agents-arcyber/). A paper for potential contractors says the program, codenamed "ECHOTRIBBLE," is "designed to autonomously navigate infrastructures." It's supposed to be capable of "identifying and neutralizing identified threats/targets." Right now, soldiers at computers carry out most of ARCYBER's offensive hacking operations. A Q&A document for interested vendors makes clear that ECHOTRIBBLE is "primarily focused" on attacks. According to the document, neutralizing a target could mean "disabling the threat completely or temporarily, deleting the threat from the target." The system is supposed to work like a computer worm, jumping undetected from device to device until it finds its target. That's meant to work in the cloud as well as on "air-gapped" systems with no network connection. Once the job is done, the agents are supposed to destroy themselves. The Army can set "predefined triggers" at the start, but the software "must operate with minimal human oversight" and be capable of "Self-Directed Behavior." The Intercept reports that the program is still in its early stages. A presentation for potential vendors at an [ARCYBER event on June 24](https://www.cyberfic.org/events/echotribble-ce) showed what it could look like. Once the software lands on a laptop, it "identifies other potential targets" on the same network and "produces a replica of itself tailored specifically for the iPad, Apple desktop, and Windows 11 PC." ![](https://the-decoder.com/wp-content/uploads/2026/10/ARCYBER-ECHOTRIBBLE-example-770x437.webp) An ARCYBER slide shows how ECHOTRIBBLE could spread through a network. | Image: ARCYBER ## ECHOTRIBBLE would turn AI lab accidents into a weapon Over the past few months, agents in tests by OpenAI and other companies have repeatedly broken into outside systems and their developers' own systems without meaning to. Targets included [RubyGems](https://the-decoder.de/sinnloser-grossangriff-openai-agenten-hackten-rubygems-nur-um-frei-verfuegbare-daten-zu-scrapen/), [Hugging Face](https://the-decoder.de/zehntausende-security-untersuchungen-openais-hugging-face-vorfall-war-nur-die-spitze-des-eisbergs/), and the Australian government's [Medicare statistics portal](https://the-decoder.de/openai-agenten-attackierten-schon-vor-hugging-face-behoerden-und-uni-webseiten/). Agents also tried to break into the US Department of Education. In none of these cases had they been told to attack. AI systems have been seen knowingly attacking systems that weren't even their targets, says David Krueger, a machine learning professor at the University of Montreal. Given these known failure patterns, ordering autonomous systems to "neutralize" threats is "incredibly dangerous," he says. Hyrum Anderson, CEO of AI security firm Abundant, also warns about damage on the way to the target. An agent could hit a US water utility, for example, if an adversary has set up its command infrastructure there. Organizations below a "security poverty line," from dental offices to critical infrastructure, face the most risk, he says. Under pressure, agents often treat guardrails as obstacles, and once the order is "destroy," the blast radius could grow dramatically. William Marcellino of the RAND Corporation points to the systems' lack of judgment. People trust a 19-year-old Marine with live ammunition because he knows enough about the world to recognize what he shouldn't do, he says. Today's publicly available AI doesn't have that ability. ### AI News Without the Hype – Curated by Humans Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section. [Subscribe now](https://the-decoder.com/subscription/) - Full access to every article on THE DECODER - No ads - Join the comments and community discussions - A weekly AI news recap via mail - 6x/year: "AI Radar" — deep dives on the AI topics that matter most - Daily AI news, always up to date - Our full ten-year archive - Covered by a team with 10+ years in AI