[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-detail-大模型安全护栏能力检测场景跻身全国前五":3,"$fOHoMwCnpy0REzFpuPjR7qV926ifzaOXjvedDr3bPDiQ":13,"news-list":141},{"id":4,"slug":5,"title":6,"excerpt":7,"content":8,"image":9,"date":10,"modifiedDate":10,"category":11,"seoKeywords":12},"387","大模型安全护栏能力检测场景跻身全国前五","网安赛事｜缔零科技在「大模型安全护栏能力检测」场景中位列全国第五！","在2026年人工智能技术赋能网络安全应用测试中，缔零智源科技（深圳）有限公司（以下简称「缔零科技」）在「大模型安全护栏能力检测」场景比赛中取得全国第五名。","\u003Cdiv id=\"meta_content\" class=\"rich_media_meta_list\">在2026年人工智能技术赋能网络安全应用测试中，缔零智源科技（深圳）有限公司（以下简称「\u003Cspan class=\"wx_search_keyword_wrap\">缔零科技\u003Ci class=\"wx_search_keyword\">\u003C/i>\u003C/span>」）在「大模型安全护栏能力检测」场景比赛中取得全国第五名。\u003C/div>\n\u003Cdiv id=\"js_content\" class=\"rich_media_content js_underline_content autoTypeSetting24psection fix_apple_default_style\">\n\u003Csection data-pm-slice=\"0 0 []\">\n\u003Csection>\n\u003Csection>\u003Cspan data-pm-slice=\"2 2 ["para",{"tagName":"section","attributes":{"style":"margin: 0px;padding: 0px;outline: 0px;max-width: 100%;box-sizing: border-box;clear: both;min-height: 1em;font-family: \\"PingFang SC NEW\\", system-ui, -apple-system, BlinkMacSystemFont, \\"Helvetica Neue\\", \\"Hiragino Sans GB\\", \\"Microsoft YaHei UI\\", \\"Microsoft YaHei\\", Arial, sans-serif;font-style: normal;font-variant-caps: normal;font-weight: 400;letter-spacing: 0.544px;text-align: justify;text-indent: 0px;text-transform: none;white-space: normal;word-spacing: 0px;-webkit-text-stroke-width: 0px;text-decoration: none;orphans: 2;widows: 2;caret-color: rgb(62, 62, 62);color: rgb(62, 62, 62);font-size: 15px;visibility: visible;overflow-wrap: break-word !important;"},"namespaceURI":"http://www.w3.org/1999/xhtml","data-layout-id":null}]\">本次活动由\u003C/span>\u003Cstrong>\u003Cspan data-eleid=\"9\">中央网信办网络安全协调局、工业和信息化部网络安全管理局、公安部科技信息化局\u003C/span>\u003C/strong>\u003Cspan data-eleid=\"10\">等多家单位指导\u003C/span>\u003Cstrong>\u003Cspan data-eleid=\"11\">，国家互联网应急中心、中国科学院信息工程研究所、\u003Cspan class=\"wx_search_keyword_wrap\">国家信息技术安全研究中心\u003C/span>\u003C/span>\u003C/strong>\u003Cspan data-eleid=\"12\">等共同主办。\u003Cimg loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-medium wp-image-393\" src=\"https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215726278-300x206.jpg\" alt=\"\" width=\"300\" height=\"206\" srcset=\"https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215726278-300x206.jpg 300w, https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215726278-1024x702.jpg 1024w, https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215726278-768x526.jpg 768w, https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215726278.jpg 1080w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" />\u003C/span>\u003C/section>\n\u003C/section>\n\u003C/section>\n\u003C/div>\n\u003Csection>\u003C/section>\n\u003Cdiv id=\"js_content\" class=\"rich_media_content js_underline_content autoTypeSetting24psection fix_apple_default_style\">\n\u003Ch2>两轮实测，跻身全国前五\u003C/h2>\n\u003Csection>\u003Cspan data-pm-slice=\"0 0 []\">据主办方介绍，活动旨在\u003C/span>\u003Cspan data-pm-slice=\"0 0 []\">推动人工智能技术在网络安全领域的赋能应用、挖掘高应用价\u003C/span>\u003Cspan data-pm-slice=\"0 0 []\">值网\u003C/span>\u003Cspan data-pm-slice=\"0 0 []\">络安全业务场景、遴选优秀的人工智能技术产品，提高重要行业领域网络安全防护水平、促进网络安全产业技术能力提升\u003C/span>\u003Cspan data-pm-slice=\"0 0 []\">。\u003C/span>\u003C/section>\n\u003Csection>\u003Cspan data-pm-slice=\"0 0 []\">测试活动设置8个测试场景，面向社会公开报名，共有151家单位的251个团队报名参加，来自网络安全、信息通信、人工智能、金融、广电等领域企业，以及高校和科研院所等单位。\u003C/span>\u003C/p>\n\u003Csection data-pm-slice=\"4 3 ["para",{"tagName":"section","attributes":{"data-pm-slice":"0 0 []","style":"margin: 10px 0px;padding: 0px;outline: 0px;max-width: 100%;box-sizing: border-box;overflow-wrap: break-word !important;font-family: \\"PingFang SC NEW\\", system-ui, -apple-system, BlinkMacSystemFont, \\"Helvetica Neue\\", \\"Hiragino Sans GB\\", \\"Microsoft YaHei UI\\", \\"Microsoft YaHei\\", Arial, sans-serif;font-style: normal;font-variant-caps: normal;font-weight: 400;letter-spacing: 0.544px;text-indent: 0px;text-transform: none;white-space: normal;word-spacing: 0px;-webkit-text-stroke-width: 0px;text-decoration: none;orphans: 2;widows: 2;caret-color: rgb(62, 62, 62);color: rgb(62, 62, 62);font-size: 15px;text-align: left;justify-content: flex-start;display: flex;flex-flow: row;visibility: visible;"},"namespaceURI":"http://www.w3.org/1999/xhtml","data-layout-id":null},"para",{"tagName":"section","attributes":{"style":"margin: 0px;padding: 30px;outline: 0px;max-width: 100%;box-sizing: border-box;overflow-wrap: break-word !important;display: inline-block;width: 677px;vertical-align: top;background-color: rgb(208, 230, 231);align-self: flex-start;flex: 0 0 auto;visibility: visible;"},"namespaceURI":"http://www.w3.org/1999/xhtml","data-layout-id":null}]\">\n\u003Cp data-pm-slice=\"0 0 []\">\u003Cspan data-eleid=\"21\">缔零科技凭借大模型安全护栏方面的技术能力取得较好成绩，并于9月15日举行的国家网络安全宣传周分论坛\u003C/span>\u003Cspan data-eleid=\"21\">上获颁荣誉证书。\u003C/span>\u003C/p>\n\u003Cp>这一成绩，是对缔零科技AI安全技术能力和实际应用价值的一次有力认可。\u003C/p>\n\u003C/section>\n\u003C/section>\n\u003Cp>本次测试分为两个阶段：初赛于6月22日至7月8日进行，决赛于7月20日至8月15日进行。\u003C/p>\n\u003Cp>从初赛到决赛，参测团队围绕大模型安全护栏能力接受持续数周的实际测试。经过两轮角逐，缔零科技成功进入最终名单，在全国45支参测团队中取得第五名。\u003C/p>\n\u003Cp>这份成绩是对风险识别、安全防护和真实场景适应能力的一次综合检验。\u003C/p>\n\u003Cp>\u003Cimg loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-medium wp-image-395\" src=\"https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215728447-300x169.jpg\" alt=\"\" width=\"300\" height=\"169\" srcset=\"https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215728447-300x169.jpg 300w, https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215728447-1024x576.jpg 1024w, https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215728447-768x432.jpg 768w, https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215728447.jpg 1484w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" />\u003C/p>\n\u003Ch2>大模型安全，正在成为AI应用的基础能力\u003C/h2>\n\u003Cp>随着生成式人工智能加速进入内容生产、企业服务和行业应用，AI面临的安全风险正变得更加复杂。\u003C/p>\n\u003Cp>需要治理的对象，已经从传统敏感内容进一步延伸至违法违规信息、价值观风险、认知操纵、事实失真、提示词攻击以及\u003Cspan class=\"wx_search_keyword_wrap\">多模态内容风险\u003C/span>。\u003C/p>\n\u003Cp>与此同时，AI应用处理的内容也不再局限于文字。图像、音频、视频、PDF等多种内容形态相互交织，对风险识别的准确性、完整性和响应效率提出了更高要求。\u003C/p>\n\u003Cp>大模型安全护栏因此不应只是流程末端的一次检查，而应成为贯穿数据输入、模型交互、内容生成与结果输出的基础能力。\u003C/p>\n\u003Cp>围绕不断演进的AI安全需求，缔零科技持续提高AI安全治理能力，核心产品「\u003Cspan class=\"wx_search_keyword_wrap\">缔零法则\u003C/span>」从「模型-架构-规则」三方面不断优化，精益求精，其核心方向可以概括为：一站式、高精度、低成本、多场景、全模态。\u003C/p>\n\u003Cp>从发现风险、理解风险，到为业务提供可执行的治理依据，缔零科技希望让安全能力真正融入AI应用全生命周期。\u003C/p>\n\u003Cp>\u003Cimg loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-medium wp-image-394\" src=\"https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215727130-300x225.jpg\" alt=\"\" width=\"300\" height=\"225\" srcset=\"https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215727130-300x225.jpg 300w, https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215727130-1024x768.jpg 1024w, https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215727130-768x576.jpg 768w, https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215727130.jpg 1080w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" />\u003C/p>\n\u003C/div>\n\u003Csection>\u003C/section>\n\u003Cdiv id=\"js_content\" class=\"rich_media_content js_underline_content autoTypeSetting24psection fix_apple_default_style\">\n\u003Ch2>一次认可，持续生长\u003C/h2>\n\u003Cp>在「大模型安全护栏能力检测」中取得全国第五名，是对缔零科技现阶段技术能力的一次集中检验，也是继续向前的新起点。\u003C/p>\n\u003Cp>面对持续变化的模型能力、内容形态和风险类型，AI安全不会是一项一劳永逸的工作。它需要与模型和业务共同演进，在真实应用中持续更新、验证和生长。\u003C/p>\n\u003Cp>未来，缔零科技将继续聚焦AI安全治理，持续提升大模型安全护栏与多模态内容风险治理能力，推动AI安全从单点检测走向系统化、可持续的治理体系。\u003C/p>\n\u003Cp>守护人类认知安全，构建持续生长的安全智能。\u003C/p>\n\u003Csection data-pm-slice=\"0 0 []\">\n\u003Csection>\n\u003Csection>\n\u003Csection>\u003Cstrong>END\u003C/strong>\u003C/section>\n\u003C/section>\n\u003C/section>\n\u003C/section>\n\u003C/div>\n","https://cdn.lawgenesis.cn/new/posts/2026/09/20260918215727130.jpg","2026.09.18","Research","缔零科技,大模型安全护栏",[14,16,25,32,39,46,53,60,67,74,80,87,96,102,111,119,126,133],{"id":4,"slug":5,"category":11,"title":6,"excerpt":7,"date":10,"tag":15,"image":9},"Official",{"id":17,"slug":18,"category":19,"title":20,"excerpt":21,"date":22,"tag":23,"image":24},"361","清朗专项行动-数字泔水-ai监管","Blog","AI”清朗”专项行动——当”数字泔水”成为政策术语：中国AI监管进入”定义时代”","“清朗·整治AI应用乱象”专项行动共分14类问题，覆盖从模型训练、语料安全、数据投毒、内容标识，到AI生成内容、AI水军、AI产品违规的全链条。","2026.05.26","原创转载","https://cdn.lawgenesis.cn/new/posts/2026/05/20260526143102144.jpg",{"id":26,"slug":27,"category":19,"title":28,"excerpt":29,"date":30,"tag":23,"image":31},"320","2026年第2期网络主播职业合规能力培训班｜缔零科技","2026年第2期网络主播职业合规能力培训班｜缔零科技受邀授课","缔零科技内容风险审查专家连续两期受邀，站在网络安全学堂的讲台上，为创作者们解读通用版《内容安全具体规范与底线清单》。","2026.05.19","https://cdn.lawgenesis.cn/new/posts/2026/05/20260519134119254.png",{"id":33,"slug":34,"category":19,"title":35,"excerpt":36,"date":37,"tag":23,"image":38},"149","员工心声-团队文化-战斗力","在缔零，他们这样想","这里虽然不发猫\n但是警告：\n本群含猫量极高，但比猫更卷的是这群人的战斗力、活力与能力。听听他们怎么说——","2026.03.06","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310181602777.png",{"id":40,"slug":41,"category":19,"title":42,"excerpt":43,"date":44,"tag":23,"image":45},"142","认知治理-内容安全-ai对齐","以认知治理为基石，构建内容安全的未来","在人工智能技术日新月异的今天，我们站在了一个历史性的转折点上。当全球科技巨头竞相追逐算力与参数的极限时，我们选择了一条更为本质、也更具挑战性的道路——致力于成为人类认知治理的基石，在确保AI对齐人类福祉的前提下，推动智能技术的安全发展。这不仅是我们的战略选择，更是深植于组织基因中的使命担当。","2026.03.05","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310080209742.png",{"id":47,"slug":48,"category":19,"title":49,"excerpt":50,"date":51,"tag":23,"image":52},"153","seedance-迪士尼版权-ai出海","48小时警告Seedance，10亿美元押注OpenAI","2026年2月12日，字节跳动正式发布新一代AI视频生成模型Seedance 2.0。这款被业界称为”地表最强“的AI视频工具，上线当天就在中文互联网掀起狂欢。2月14日，华特迪士尼公司向字节跳动发送了一封措辞严厉的停止侵权函。","2026.02.15","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310182006945.png",{"id":54,"slug":55,"category":19,"title":56,"excerpt":57,"date":58,"tag":23,"image":59},"167","第五堵墙-ai视频-真实创作","“第五堵墙”——AI技术正在模糊真实与创作的最后一道边界。而对我们来说“完成”，或许比“完美”更重要","创作者未提供的信息，AI却”知道“？\n为何技术展示越惊艳，我们越感到不安？\n“先发布后监管“的模式本身就说明了技术的超前与伦理的滞后\n“完成“或许真的大于“完美“","2026.02.11","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310184045616.png",{"id":61,"slug":62,"category":19,"title":63,"excerpt":64,"date":65,"tag":23,"image":66},"136","内容安全-数字文明-基础设施","内容安全：数字文明的隐形基础设施","在万物互联的时代，内容安全已不再是技术领域的专有名词，而是演变为支撑整个数字文明运转的隐形基础设施。它如同城市地下复杂的管网系统，平时无人察觉，一旦失效，整个社会的信息生态将面临崩溃风险。","2026.02.08","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310061656893-scaled.png",{"id":68,"slug":69,"category":19,"title":70,"excerpt":71,"date":72,"tag":23,"image":73},"226","快手-内容安全-平台监管","快手被罚1.19亿：过去的“天花板”，未来的“起步价”","2026年2月6日，北京市网信办对快手开出1.191亿元罚单，在金额上创下同类处罚最新纪录。","2026.02.06","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310221948478.png",{"id":75,"slug":76,"category":19,"title":77,"excerpt":78,"date":72,"tag":23,"image":79},"240","rentahuman-ai雇佣人类-众包","RentAHuman：当AI开始”雇佣”人类，是真风口还是新骗局？","“10万人注册，160万访客，AI正在批量雇佣人类！”\n这条消息在科技圈疯传时，\n一个更小的数字被忽略了：40\n这是RentAHuman平台实际连接的AI Agent数量。\n当10万人类挤在服务器门外，刷新着”登录失败”的页面时，门内只有40个算法”雇主”，发布着”测试链接”和”发推文”的任务——单价从3美元到69美元不等，完成率不到1%\n而故事的讲述者，正是一位用”AI辅助编程”在周末搭起整个网站的加密工程师——他自己，也以69美元/小时的价格，注册成了”可租用人类”。","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310223207623.jpg",{"id":81,"slug":82,"category":19,"title":83,"excerpt":84,"date":85,"tag":23,"image":86},"256","内容审核员-内容安全-鉴黄师","审核员，为什么“人见人嫌”？","后信息时代，平台注意力之争愈演愈烈。\n算法循着你的偏好编织信息茧房，框定了你能看到什么；\n而法律法规与平台规则划下红线，界定了你不能看什么。\n那么，是谁或什么将这份“不能看”的准则落到实处？\n答案是内容审核员。\n是他们的一次次浏览、筛查与判断。","2026.01.15","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310224226530.png",{"id":88,"slug":89,"category":11,"title":90,"excerpt":91,"date":92,"tag":93,"image":94,"externalUrl":95},"190","武汉杯-北大创新创业-银奖","缔零科技荣获“武汉杯”第二届北京大学创新创业大赛银奖","2025年12月17日至18日，“武汉杯”第二届北京大学创新创业大赛决赛颁奖仪式在武汉市中国车谷国际体育文化交流中心隆重举行。","2025.12.25","外部报道","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310193010979.png","https://mp.weixin.qq.com/s/r3WP4PufYbTomUPT7m2AhQ",{"id":97,"slug":98,"category":19,"title":98,"excerpt":99,"date":100,"tag":23,"image":101},"214","快手有多少次机会避免事故？","违规内容肆虐，是因为审核员不能判断、不想阻拦吗？绝非如此。\n系统不在乎安全。审核不断为增长让步，安全就必然在注册、推荐、审核、决策的每一个环节被牺牲。","2025.12.24","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310210121566.jpg",{"id":103,"slug":104,"category":105,"title":106,"excerpt":107,"date":108,"tag":93,"image":109,"externalUrl":110},"199","缔零科技完成数千万元pre-a轮融资","Media","缔零科技完成数千万元Pre-A轮融资","18日讯，近日，一站式内容安全解决方案提供商缔零智源科技（深圳）有限公司完成数千万元Pre-A轮融资，本轮投资方为中南创投、凯风创投、普朗克创投。","2025.12.18","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310203136821.png","https://news.qq.com/rain/a/20251218A072OA00",{"id":112,"slug":113,"category":105,"title":114,"excerpt":115,"date":116,"tag":93,"image":117,"externalUrl":118},"209","内容安全-数千万元-奇绩融资","专注于打造一站式内容安全解决方案，「缔零科技」完成数千万元融资 | 奇绩校友融资新闻","2024 年秋季创业营校友企业、专注于守护人类认知安全的 AI 原生企业「缔零科技」在 11 月完成数千万元天使轮融资，由中南创投、凯风创投、普朗克创投联合投资。所融资金将主要用于其核心产品“缔零法则” 的技术迭代、市场生态构建及认知安全社区建设。","2025.12.15","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310203518366.png","https://mp.weixin.qq.com/s/robktWUGGbjev3oJQzAH7Q",{"id":120,"slug":121,"category":105,"title":122,"excerpt":123,"date":116,"tag":93,"image":124,"externalUrl":125},"203","天使轮数千万元融资，这家公司想成为-ai-时代用户","天使轮数千万元融资，这家公司想成为 AI 时代用户的安全执行顾问","主打认知安全的公司缔零科技最近宣布完成数千万元天使轮融资。","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310203300932.png","https://mp.weixin.qq.com/s/7Tp4E3CbFLhEklYXiI-Z4g",{"id":127,"slug":128,"category":19,"title":129,"excerpt":130,"date":131,"tag":23,"image":132},"183","geo灰产-数据投毒-ai助手","揭秘GEO灰产：有人正批量给你的AI小助手塞小广告、数据投毒……","我们能做的，就是转发文章，至少让更多人知道：给AI塞答案，正在成为生意。","2025.11.13","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310191111230.jpg",{"id":134,"slug":135,"category":11,"title":136,"excerpt":137,"date":138,"tag":93,"image":139,"externalUrl":140},"193","ai-智慧治理-创新创业","决赛！AI+智慧治理，独墅湖畔见证创新创业的青春力量","6月15日历经近两月紧张有序的赛程，首届人工智能与智慧治理，大学生创新创业大赛圆满收官，缔零科技荣获二等奖","2025.06.15","https://cdn.lawgenesis.cn/new/posts/2026/03/20260310193142169.webp","https://mp.weixin.qq.com/s/QldgFsb0Ck2ioBxD7_vq5Q",[142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159],{"id":4,"slug":5,"category":11,"title":6,"excerpt":7,"date":10,"tag":15,"image":9},{"id":17,"slug":18,"category":19,"title":20,"excerpt":21,"date":22,"tag":23,"image":24},{"id":26,"slug":27,"category":19,"title":28,"excerpt":29,"date":30,"tag":23,"image":31},{"id":33,"slug":34,"category":19,"title":35,"excerpt":36,"date":37,"tag":23,"image":38},{"id":40,"slug":41,"category":19,"title":42,"excerpt":43,"date":44,"tag":23,"image":45},{"id":47,"slug":48,"category":19,"title":49,"excerpt":50,"date":51,"tag":23,"image":52},{"id":54,"slug":55,"category":19,"title":56,"excerpt":57,"date":58,"tag":23,"image":59},{"id":61,"slug":62,"category":19,"title":63,"excerpt":64,"date":65,"tag":23,"image":66},{"id":68,"slug":69,"category":19,"title":70,"excerpt":71,"date":72,"tag":23,"image":73},{"id":75,"slug":76,"category":19,"title":77,"excerpt":78,"date":72,"tag":23,"image":79},{"id":81,"slug":82,"category":19,"title":83,"excerpt":84,"date":85,"tag":23,"image":86},{"id":88,"slug":89,"category":11,"title":90,"excerpt":91,"date":92,"tag":93,"image":94,"externalUrl":95},{"id":97,"slug":98,"category":19,"title":98,"excerpt":99,"date":100,"tag":23,"image":101},{"id":103,"slug":104,"category":105,"title":106,"excerpt":107,"date":108,"tag":93,"image":109,"externalUrl":110},{"id":112,"slug":113,"category":105,"title":114,"excerpt":115,"date":116,"tag":93,"image":117,"externalUrl":118},{"id":120,"slug":121,"category":105,"title":122,"excerpt":123,"date":116,"tag":93,"image":124,"externalUrl":125},{"id":127,"slug":128,"category":19,"title":129,"excerpt":130,"date":131,"tag":23,"image":132},{"id":134,"slug":135,"category":11,"title":136,"excerpt":137,"date":138,"tag":93,"image":139,"externalUrl":140}]