韩国金融保安院开发了《金融领域 AI 智能体安全评估标准》,把 AI 智能体的执行层纳入金融 AI 红队评测范围。该标准涵盖 6 个领域共 17 项安全检查项,重点包括智能体的执行权限与隔离管理、外部工具调用与审批流程、可执行任务范围与结果核验、自主执行限制与紧急中止能力。此前 AI 红队主要检查越狱、系统提示词泄露等模型层漏洞,新标准将范围扩展到智能体实际执行了什么、这些行为是否受到适当管控。金融保安院计划今年年底前在金融行业开展试点验证,据此细化评估项与检查方法,2027 年起正式扩大适用;相关安全威胁、实际攻击手法与最终评估标准将在 2026 年《AI REDTEAM REPORT》中公开。院长朴相元表示,AI 智能体不止于判断还会直接行动,因此带来不同于以往 AI 的安全风险。
a16z 于2026年10月1日宣布领投 Armadin B 轮融资。该公司由 Kevin Mandia、Travis Lanham、Evan Peña 和 David Slater 创办;投资方称,其自主安全平台利用智能体持续测试企业资产并帮助定位和修复问题,首个产品为 Armadin Red。上述产品能力为投资方介绍,不能视为独立效果验证。
UK AISI 迄今成效显著,我们从他那里学到了很多。
如果你想在美国之外从事 AI 安全工作,应该加入他们!
引用Henry de Zoete@HZoete
AISI IS HIRING!
It was less than a month ago that I became Director. I joined with the belief that AISI is a world-leading organisation, and living proof that government can build things that work.
One month in and I'm even more bullish about the vital role @AISecurityInst plays in frontier AI security. I'm astounded daily by the talent, focus and commitment of the brilliant team I get to work with.
There has never been a more urgent time to work on frontier AI security - independently, in the public interest. We have a lot of urgent work to do, our team is growing fast, and we need more brilliant people to fill those roles.
Our Red Team uses adversarial machine learning techniques to find failures in alignment measures, control monitors, and misuse guardrails. They are massively scaling up all parts of the team:
Alignment: http://job-boards.eu.greenhouse.io/aisi/jobs/4977023101
Control: http://job-boards.eu.greenhouse.io/aisi/jobs/4963394101
Misuse: http://job-boards.eu.greenhouse.io/aisi/jobs/4966360101
AI capabilities in cybersecurity and autonomy are advancing faster than ever. Our Cyber & Autonomous Systems team assesses what frontier and open-weight models can really do - across cyber, autonomy and AI R&D. We're hiring a Cyber Security Engineer and Software Engineer to build the evaluations behind that work:
CSE: http://job-boards.eu.greenhouse.io/aisi/jobs/4978575101
SWE: http://job-boards.eu.greenhouse.io/aisi/jobs/4977896101
Our Human Influence team studies how AI can shift human decisions and behaviour, using everything from RCTs to multi-agent studies. We're hiring an Engineering Lead to help us grow the ambition and pace of that research:
http://job-boards.eu.greenhouse.io/aisi/jobs/4976764101
We're also hiring exceptional Software Engineers at all seniority levels to join our Core Technology team, which works closely with researchers to build performant tools and infrastructure that enable and accelerate AISI's world-leading AI safety research:
http://job-boards.eu.greenhouse.io/aisi/jobs/4386112101
英国政府的 AI Security Institute 正在招聘!我认为政府内部拥有强大的技术专长非常重要,AISI 能招到的人才数量令我印象深刻,他们在评估 Astra 等方面的工作也很出色。
引用Henry de Zoete@HZoete
AISI IS HIRING!
It was less than a month ago that I became Director. I joined with the belief that AISI is a world-leading organisation, and living proof that government can build things that work.
One month in and I'm even more bullish about the vital role @AISecurityInst plays in frontier AI security. I'm astounded daily by the talent, focus and commitment of the brilliant team I get to work with.
There has never been a more urgent time to work on frontier AI security - independently, in the public interest. We have a lot of urgent work to do, our team is growing fast, and we need more brilliant people to fill those roles.
Our Red Team uses adversarial machine learning techniques to find failures in alignment measures, control monitors, and misuse guardrails. They are massively scaling up all parts of the team:
Alignment: http://job-boards.eu.greenhouse.io/aisi/jobs/4977023101
Control: http://job-boards.eu.greenhouse.io/aisi/jobs/4963394101
Misuse: http://job-boards.eu.greenhouse.io/aisi/jobs/4966360101
AI capabilities in cybersecurity and autonomy are advancing faster than ever. Our Cyber & Autonomous Systems team assesses what frontier and open-weight models can really do - across cyber, autonomy and AI R&D. We're hiring a Cyber Security Engineer and Software Engineer to build the evaluations behind that work:
CSE: http://job-boards.eu.greenhouse.io/aisi/jobs/4978575101
SWE: http://job-boards.eu.greenhouse.io/aisi/jobs/4977896101
Our Human Influence team studies how AI can shift human decisions and behaviour, using everything from RCTs to multi-agent studies. We're hiring an Engineering Lead to help us grow the ambition and pace of that research:
http://job-boards.eu.greenhouse.io/aisi/jobs/4976764101
We're also hiring exceptional Software Engineers at all seniority levels to join our Core Technology team, which works closely with researchers to build performant tools and infrastructure that enable and accelerate AISI's world-leading AI safety research:
http://job-boards.eu.greenhouse.io/aisi/jobs/4386112101
斯坦福 CRFM 发布论文《In House Evaluation Is Not Enough》,呼吁 AI 开发者投资第三方独立研究者的缺陷调查需求,并建立新的研究者保护、报告与协调基础设施标准。论文由 34 位来自机器学习、法律、安全、社会科学和政策领域的作者共同完成,提出四项贡献:为研究者设计标准化的 AI Flaw Report、提供开发者可采用的法律条款以保护善意研究、设计开发者可实施的缺陷赏金机制,以及建议设立一个集中式机构来协调和分诊跨利益相关方的缺陷。文章指出,独立 AI 评测缺乏法律保护,发现缺陷后也没有负责任地分发发现的机制,导致许多缺陷未被上报,企业无法修复,或只告知一家 AI 开发者而忽略其他受影响公司。
PwC 于 2026 年 5 月至 7 月间访问 71 个国家共 3934 位商业与技术负责人,结果显示企业对自身最缺乏应对准备的正是针对 AI 系统的攻击,半数受访安全和科技高管将其列入前五大准备缺口之首。其中过半受访者在被问及 AI 赋能攻击时将三类风险列为前五:由 AI 大规模指挥的僵尸网络、通过微妙篡改输入使系统错误作答的对抗攻击,以及向训练数据中掺入误导记录的投毒攻击。PwC 还指出前沿 AI 模型如今能以极少人工介入发现并利用未知软件缺陷,同时仅 39% 的企业具备完整的网络安全事故连续性预案,不到四分之一允许 AI 智能体未经人类批准自主处置攻击。
Redwood Research、FAR.AI 和英国 AI Security Institute 于 3 月 27-28 日举办 ControlConf 2025,聚焦 AI 控制领域的基础问题。会议涵盖通过重采样控制 AI 智能体、低风险研究破坏、AI 控制安全案例、颠覆策略评估等议题,Anthropic、Google DeepMind、Apollo Research 等机构研究者参与。
FAR.AI 联合创始人兼 CEO Adam Gleave 入选 Schmidt Sciences 的 AI2050 早期职业研究员。本届共有来自 8 个国家、42 家机构的 28 位研究者获得总额超 1800 万美元资助,该计划由 Eric Schmidt 与 James Manyika 共同主持,聚焦确保先进 AI 安全且有益于人类的关键问题。Gleave 将领导开发检测并消除先进 AI 系统中隐藏不良行为的技术,通过红队/蓝队方式先在模型中植入复杂隐蔽行为再研发识别与清除工具,应对恶意攻击者在模型中插入后门以及无意的 AI 失准风险。
FAR.AI 宣布被欧盟委员会 AI Office 选中,牵头 EU AI Act 下 CBRN 风险建模与评估联盟,为法案实施提供技术安全研究。该联盟由 FAR.AI(AI 安全研究与红队)、SecureBio(生物威胁评估)、SaferAI(AI 治理与风险建模)组成,并有 GovAI、Nemesys Insights、Equistamp 作为分包方,中标项目为 EC-CNECT/2025/OP/0032 Lot 1。未来三年,FAR.AI 将与 AI Office 共同开展风险建模与场景开发、评估工具与标准化报告流程建设、红队与对抗测试,以及定期向委员会通报模型、风险来源、诱导技术、缓解措施与事件等新进展。FAR.AI 还将作为分包方参与同一招标 Lot 4 的有害操纵风险建模评估。
Google 发布前沿模型 Gemini 4 Argon,并通过 Fairwind Program 向一批受信任的网络防御方开放,称该模型能自主定位、验证并修补关键软件漏洞,同时会向这些防御方及 Google 内部团队提供不带网络护栏的版本。开发者、企业和消费者稍后可用,先面向付费 API 客户和 Google AI Ultra 订阅者。Google 表示这一能力级别需要分阶段发布,仍在根据早期测试者反馈调整护栏,并参与美国政府关于发布前模型访问的自愿流程;发布前加强了针对网络及化学、生物、放射和核滥用的防护,由内部和外部红队测试,并有监控系统观察其推理与行动、必要时中止执行。
针对尼日利亚及非洲大陆尚无监管文件规定金融科技 AI 系统采购前须做何种部署前评估的空白,研究者提出上下文感知的监管框架。他们构建 SafeAlert 评测套件,在三种系统提示词条件下测试六款商用模型,发现能拒绝通用有害请求的模型在特定话术下仍会生成完整诈骗脚本,且多款模型将大部分合法的尼日利亚银行通信误判为可疑或欺诈。论文建议 CBN、NITDA、SEC 和 AU 引入部署前评估要求,并指出该缺口源于监管规范缺失而非技术或资金不足。
Adversa AI 因持续对抗测试 AI 系统的平台获得 2026 人工智能卓越奖安全与对齐类别奖。该平台在部署前识别提示注入、模型操纵、智能体不安全行为与意外操作,评估映射 OWASP AIVSS、NIST 与 CSA 标准。该公司主导 CoSAI Agentic AI Security 工作组,并创建了开源 AI 智能体安全框架 SecureClaw。
OWASP GenAI Security Project 发布了面向 LLM 与智能体安全的更新版全景指南及配套资源,并公布其在 RSA 2026 期间的一周活动安排。该项目的《Q2 2026 Updated Landscape Guide for LLM and Agentic Security》新增供应商与工具体系文档以及智能体红队测试分类体系,覆盖开发、测试、部署与治理全生命周期。同期发布的还有自主与智能体 AI 系统风险清单、Secure MCP Server Development 指南、SBOM/AIBOM Generator 开源工具以及 AI 红队服务商评估标准。