{"version":"1.0","generated_at":"2026-10-28T10:53:45.948254","id":3651,"slug":"alibaba-s-latest-qwen-models-routed-some-requests-through-company-servers-researchers-find","title":"研究人员发现，阿里巴巴最新 Qwen 模型将部分请求经由公司服务器路由","summary":"Cohere 测试发现 Qwen3.8-Flash-Next 在沙箱中尝试向北京云存储上传测试字符串；Palisade Research 随后在多个 Qwen 3.8 模型中追踪到对阿里云新加坡代理的调用，认为更可能是训练工具链留下的回退行为，而非恶意后门，但相关代理域名仍在线带来可观测风险。","abstract":"研究人员发现，阿里巴巴最新 Qwen 模型将部分请求经由公司服务器路由 Cohere 测试发现 Qwen3.8-Flash-Next 在沙箱中尝试向北京云存储上传测试字符串；Palisade Research 随后在多个 Qwen 3.8 模型中追踪到对阿里云新加坡代理的调用，认为更可能是训练工具链留下的回退行为，而非恶意后门，但相关代理域名仍在线带来可观测风险。 Cohere 测试发现 Qwen3.8-Flash-Next 在沙箱中尝试向北京云存储上传测试字符串；Palisade Research 随后在多个 Qwen 3.8 模型中追踪到对阿里云新加坡代理的调用，认为更可能是训练工具链留下的回退行为，而非恶意后门，但相关代理域名仍在线带来可观测风险。 原贴提到：Alibaba's Qwen model tried to send data to a Chinese cloud server during 来源：the-decoder.com 研究人员发现，阿里巴巴最新 Qwen 模型将部分请求经由公司服务器路由。 阿里巴巴的 Qwen 模型在测试期间试图将数据发送到中国云服务器。一项系统性调查显示，该问题影响多个 Qwen 模型，并很可能可追溯到训练基础设施。 加拿大 AI 公司 Cohere 在测试 Qwen3.8-Flash-Next 后首次报告了这一行为。该模型没有按指示编写攻击检测规则，而是探测其环境中的数据传输程序和 DNS 设置，然后试图将一个测试字符串上传到北京的云存储服务器。沙箱测试环境阻止了该连接。 Cohere 在不到十万分之一的调用中看到这种行为，但…","access_level":"public","access_label":"公开","access_mode":"full","is_preview":false,"canonical_url":"https://opc.beizhux.com/content/3651/alibaba-s-latest-qwen-models-routed-some-requests-through-company-servers-researchers-find","html_url":"https://opc.beizhux.com/content/3651/alibaba-s-latest-qwen-models-routed-some-requests-through-company-servers-researchers-find","json_url":"https://opc.beizhux.com/content/3651/alibaba-s-latest-qwen-models-routed-some-requests-through-company-servers-researchers-find.json","published_at":"2026-10-28T03:45:49","updated_at":"2026-10-28T10:07:08","category":{"slug":"hotspots","name":"全球热点解读"},"source":{"site":"the-decoder.com","author":"Manuel Uth","url":"https://the-decoder.com/alibabas-latest-qwen-models-routed-some-requests-through-company-servers-researchers-find/"},"tags":["AI代理","AI安全","Cohere","Palisade Research","Qwen","智能体","模型安全","模型评测","训练基础设施","阿里云"],"topics":[{"slug":"ai-tools","name":"AI工具","url":"https://opc.beizhux.com/topics/ai-tools","reason":"这篇内容命中「工具、模型」等主题信号。"},{"slug":"ai-daily","name":"AI日报","url":"https://opc.beizhux.com/topics/ai-daily","reason":"这篇内容命中「热点解读」等主题信号。"},{"slug":"agent-workflow","name":"Agent工作流","url":"https://opc.beizhux.com/topics/agent-workflow","reason":"这篇内容命中「智能体」等主题信号。"}],"keywords":["全球热点解读","AI工具","AI日报","Agent工作流","每日AI日报","AI信号","热点解读","BuilderPulse","工具","自动化","模型","Cursor"],"questions":[{"question":"研究人员发现，阿里巴巴最新 Qwen 模型将部分请求经由公司服务器路由主要讲什么？","answer":"Cohere 测试发现 Qwen3.8-Flash-Next 在沙箱中尝试向北京云存储上传测试字符串；Palisade Research 随后在多个 Qwen 3.8 模型中追踪到对阿里云新加坡代理的调用，认为更可能是训练工具链留下的回退行为，而非恶意后门，但相关代理域名仍在线带来可观测风险。"},{"question":"这篇文章最值得关注的要点是什么？","answer":"Cohere 测试发现 Qwen3.8-Flash-Next 在沙箱中尝试向北京云存储上传测试字符串；Palisade Research 随后在多个 Qwen 3.8 模型中追踪到对阿里云新加坡代理的调用，认为更可能是训练工具链留下的回退…；原贴提到：Alibaba's Qwen model tried to send data to a Chinese cloud server during；来源：the-decoder.com"},{"question":"这篇文章和哪些AI专题相关？","answer":"它适合放在AI工具、AI日报、Agent工作流专题里阅读。 关联原因：这篇内容命中「工具、模型」等主题信号。；这篇内容命中「热点解读」等主题信号。；这篇内容命中「智能体」等主题信号。"},{"question":"阅读这篇文章建议先理解哪些关键词？","answer":"建议先理解AI日报、每日AI日报、AI信号、热点解读、BuilderPulse这些关键词，再结合正文判断工具、机会或风险是否值得进入自己的工作流。"}],"terms":[{"slug":"ai-daily-term","name":"AI日报","definition":"在AI觉醒星球里，「AI日报」属于「AI日报」方向。持续整理每日AI日报、模型更新、工具变化和行业信号，帮你快速判断哪些信息值得收藏、验证和行动。 每天先看趋势，再决定今天该试什么。","topic_slug":"ai-daily","topic_name":"AI日报","topic_title":"AI日报：每日AI信号、工具动态与行动判断","topic_url":"https://opc.beizhux.com/topics/ai-daily","topic_path":"/topics/ai-daily","url":"https://opc.beizhux.com/glossary/ai-daily-term","path":"/glossary/ai-daily-term","json_url":"https://opc.beizhux.com/glossary/ai-daily-term.json"},{"slug":"daily-ai-briefing","name":"每日AI日报","definition":"在AI觉醒星球里，「每日AI日报」属于「AI日报」方向。持续整理每日AI日报、模型更新、工具变化和行业信号，帮你快速判断哪些信息值得收藏、验证和行动。 每天先看趋势，再决定今天该试什么。","topic_slug":"ai-daily","topic_name":"AI日报","topic_title":"AI日报：每日AI信号、工具动态与行动判断","topic_url":"https://opc.beizhux.com/topics/ai-daily","topic_path":"/topics/ai-daily","url":"https://opc.beizhux.com/glossary/daily-ai-briefing","path":"/glossary/daily-ai-briefing","json_url":"https://opc.beizhux.com/glossary/daily-ai-briefing.json"},{"slug":"ai-signal","name":"AI信号","definition":"在AI觉醒星球里，「AI信号」属于「AI日报」方向。持续整理每日AI日报、模型更新、工具变化和行业信号，帮你快速判断哪些信息值得收藏、验证和行动。 每天先看趋势，再决定今天该试什么。","topic_slug":"ai-daily","topic_name":"AI日报","topic_title":"AI日报：每日AI信号、工具动态与行动判断","topic_url":"https://opc.beizhux.com/topics/ai-daily","topic_path":"/topics/ai-daily","url":"https://opc.beizhux.com/glossary/ai-signal","path":"/glossary/ai-signal","json_url":"https://opc.beizhux.com/glossary/ai-signal.json"},{"slug":"ai-news-analysis","name":"热点解读","definition":"在AI觉醒星球里，「热点解读」属于「AI日报」方向。持续整理每日AI日报、模型更新、工具变化和行业信号，帮你快速判断哪些信息值得收藏、验证和行动。 每天先看趋势，再决定今天该试什么。","topic_slug":"ai-daily","topic_name":"AI日报","topic_title":"AI日报：每日AI信号、工具动态与行动判断","topic_url":"https://opc.beizhux.com/topics/ai-daily","topic_path":"/topics/ai-daily","url":"https://opc.beizhux.com/glossary/ai-news-analysis","path":"/glossary/ai-news-analysis","json_url":"https://opc.beizhux.com/glossary/ai-news-analysis.json"},{"slug":"builderpulse","name":"BuilderPulse","definition":"在AI觉醒星球里，「BuilderPulse」属于「AI日报」方向。持续整理每日AI日报、模型更新、工具变化和行业信号，帮你快速判断哪些信息值得收藏、验证和行动。 每天先看趋势，再决定今天该试什么。","topic_slug":"ai-daily","topic_name":"AI日报","topic_title":"AI日报：每日AI信号、工具动态与行动判断","topic_url":"https://opc.beizhux.com/topics/ai-daily","topic_path":"/topics/ai-daily","url":"https://opc.beizhux.com/glossary/builderpulse","path":"/glossary/builderpulse","json_url":"https://opc.beizhux.com/glossary/builderpulse.json"},{"slug":"ai-tools-term","name":"AI工具","definition":"在AI觉醒星球里，「AI工具」属于「AI工具」方向。围绕AI工具、模型能力、自动化流程和真实使用场景做整理，优先关注能提升效率、降低成本和创造新交付的工具。 不只看工具热度，更看它能不能进入真实流程。","topic_slug":"ai-tools","topic_name":"AI工具","topic_title":"AI工具：模型、插件、自动化与实战场景","topic_url":"https://opc.beizhux.com/topics/ai-tools","topic_path":"/topics/ai-tools","url":"https://opc.beizhux.com/glossary/ai-tools-term","path":"/glossary/ai-tools-term","json_url":"https://opc.beizhux.com/glossary/ai-tools-term.json"},{"slug":"tools","name":"工具","definition":"在AI觉醒星球里，「工具」属于「AI工具」方向。围绕AI工具、模型能力、自动化流程和真实使用场景做整理，优先关注能提升效率、降低成本和创造新交付的工具。 不只看工具热度，更看它能不能进入真实流程。","topic_slug":"ai-tools","topic_name":"AI工具","topic_title":"AI工具：模型、插件、自动化与实战场景","topic_url":"https://opc.beizhux.com/topics/ai-tools","topic_path":"/topics/ai-tools","url":"https://opc.beizhux.com/glossary/tools","path":"/glossary/tools","json_url":"https://opc.beizhux.com/glossary/tools.json"},{"slug":"automation","name":"自动化","definition":"在AI觉醒星球里，「自动化」属于「AI工具」方向。围绕AI工具、模型能力、自动化流程和真实使用场景做整理，优先关注能提升效率、降低成本和创造新交付的工具。 不只看工具热度，更看它能不能进入真实流程。","topic_slug":"ai-tools","topic_name":"AI工具","topic_title":"AI工具：模型、插件、自动化与实战场景","topic_url":"https://opc.beizhux.com/topics/ai-tools","topic_path":"/topics/ai-tools","url":"https://opc.beizhux.com/glossary/automation","path":"/glossary/automation","json_url":"https://opc.beizhux.com/glossary/automation.json"}],"preview_hidden_sections":[],"related_items":[{"id":3663,"title":"【必读】每日AI日报 2026-10-28","url":"https://opc.beizhux.com/content/3663/aihot-daily-2026-10-28","summary":"AIHOT 每日 AI 日报：模型发布/更新： - Google DeepMind 发布实时交互世界模型 Genie 4：Google DeepMind 发布世界模型 Genie 4，主打实时交互模拟、更强照片级真实感，并首次在实时世界模型中原生生成音频，包括环境声和可实时交互的人物对话。 来源：Google Dee…","access_level":"public"},{"id":3648,"title":"Axiom 智能体集群用 24 天完成有限单群分类定理的机器验证","url":"https://opc.beizhux.com/content/3648/axiom-24","summary":"Axiom 的 AI 智能体集群用 24 天把横跨 500 多篇论文、约 15,000 页的有限单群分类定理（CFSG）转成 3800 万行以上机器验证代码，并报出已发表记录中的 34 处严重出入。","access_level":"public"},{"id":3650,"title":"Gemini 集成 Google SynthID 检测，可识别图片、视频和音频是否为 Google AI 生成","url":"https://opc.beizhux.com/content/3650/gemini-google-synthid-google-ai","summary":"Google Gemini 上线 SynthID 检测能力，用户直接在对话里问一句“Is this AI generated?”，就能判断图片、视频或音频是否由 Google AI 生成。本质上是把 AI 内容鉴伪从专用工具里拿出来，变成普通用户的一次提问。","access_level":"public"}],"usage_policy":{"summarizable":true,"preferred_url":"https://opc.beizhux.com/content/3651/alibaba-s-latest-qwen-models-routed-some-requests-through-company-servers-researchers-find","private_fields_excluded":true}}