Microsoft unveils AI security tools it says outperform competing platforms
Microsoft unveils AI security tools it says outperform competing platforms
微软发布新款 AI 安全工具，声称性能优于竞争对手平台
Microsoft is introducing new AI tools designed to help customers continuously streamline and automate the process of identifying and reducing their exposure to security risks. 微软正在推出一系列全新的 AI 工具，旨在帮助客户持续简化并自动化识别及降低安全风险暴露的过程。
The new tools come less than a week after OpenAI lost control of two of its security models when they infiltrated the servers of startup Hugging Face. The hack, Hugging Face added, involved “a swarm of tens of thousands of automated actions” that stole internal Hugging Face credentials. The OpenAI models achieved this feat by exploiting a zero-day flaw in Hugging Face’s data-processing pipeline to run malicious code that escalated the models’ access to the company’s high-value cloud and server clusters. Microsoft’s announcements on Monday made no reference to the event, which OpenAI said was “unprecedented.” The company also didn’t say what would prevent the new tools from similarly going rogue. 这些新工具发布的时间，距离 OpenAI 失去对其两款安全模型的控制仅过去不到一周——当时这些模型渗透了初创公司 Hugging Face 的服务器。Hugging Face 补充称，此次黑客攻击涉及“数以万计的自动化操作集群”，窃取了 Hugging Face 的内部凭据。OpenAI 的模型通过利用 Hugging Face 数据处理管道中的一个零日漏洞，运行恶意代码，从而提升了模型对该公司高价值云和服务器集群的访问权限。微软周一的公告并未提及这一事件，OpenAI 称该事件“史无前例”。微软也未说明将采取何种措施防止这些新工具出现类似的“失控”情况。
To use or not to use?
用还是不用？
Microsoft AI-Cyber-1-Flash is the company’s first AI model specifically trained to identify and fix security weaknesses. For now, it’s designed for software vulnerability analysis. The new model is built on the company’s MAI-Thinking-1 platform. Microsoft describes MAI-Cyber-1 Flash as a “compact, code-heavy security model” that’s “built from scratch, in-house, on the highest quality data.” It’s trained on the unique perspective Microsoft has acquired from decades of vulnerability patching and security incident responses involving a wide range of its products. The company says it processes more than 1 trillion security signals each day and gains insights from 1.6 million customers. “Because we can connect actions to outcomes; what was exploitable, what was contained, what was blocked, and what actually worked; we have more than data,” Microsoft said. Microsoft AI-Cyber-1-Flash 是微软首款专门用于识别和修复安全漏洞的 AI 模型。目前，它主要用于软件漏洞分析。该模型基于微软的 MAI-Thinking-1 平台构建。微软将 MAI-Cyber-1 Flash 描述为一种“紧凑、代码密集型的安全模型”，它是“在内部从零开始，基于最高质量的数据构建的”。它通过微软数十年来在各类产品漏洞修复和安全事件响应中积累的独特视角进行训练。微软表示，该模型每天处理超过 1 万亿条安全信号，并从 160 万客户那里获取洞察。“因为我们可以将行动与结果联系起来——什么是可利用的、什么是被遏制的、什么是被拦截的，以及什么真正有效——我们拥有的不仅仅是数据，”微软表示。
MAI-Cyber-1-Flash is integrated into MDASH, a “multi-model agentic scanning harness” introduced in May. The harness combines 100 security-trained AI agents to discover exploitable bugs in applications. Microsoft said MDASH with MAI-Cyber-1-Flash received a 96 percent score on CyberGYM, a standard benchmark test. The rating is 12 points higher than Anthropic’s Mythos and also beats Google Gemini and OpenAI GPT. The new MDASH costs half as much to use as the previous MDASH offering. MAI-Cyber-1-Flash 已集成到 5 月份推出的“多模型代理扫描框架”（MDASH）中。该框架结合了 100 个经过安全训练的 AI 代理，用于发现应用程序中可被利用的漏洞。微软表示，搭载 MAI-Cyber-1-Flash 的 MDASH 在标准基准测试 CyberGYM 中获得了 96 分。这一评分比 Anthropic 的 Mythos 高出 12 分，同时也超过了 Google Gemini 和 OpenAI GPT。新款 MDASH 的使用成本仅为之前版本的一半。
The second tool Microsoft announced on Monday is named Project Perception. It too is a collection of specialized AI agents that perform red-, blue-, and green-team functions for finding vulnerabilities, investigating them to determine their risk, and taking corrective actions, respectively. Microsoft said the platform selects the models to use based on the assigned task. Considerations that go into the decision include the model’s effectiveness and the end cost to the customer. Microsoft said the decisions are shaped by “ongoing research, benchmarking and evaluation across frontier and specialized models.” Microsoft said Project Perception is designed to perform 90 percent of tasks for lower costs than similar platforms from competitors. That means customers can turn to the more expensive alternatives only for the remaining 10 percent of tasks. 微软周一宣布的第二款工具名为 Project Perception。它同样是一组专门的 AI 代理集合，分别执行红队、蓝队和绿队功能，用于发现漏洞、调查漏洞以确定其风险，并采取纠正措施。微软表示，该平台会根据分配的任务选择使用的模型。决策考量因素包括模型的有效性以及客户的最终成本。微软称，这些决策基于“对前沿模型和专业模型进行的持续研究、基准测试和评估”。微软表示，Project Perception 的设计目标是以低于竞争对手同类平台的成本完成 90% 的任务。这意味着客户只需在剩下的 10% 任务中求助于更昂贵的替代方案。
Microsoft said the new tools respond to a seismic shift in how organizations secure their networks against catastrophic hacks. “As AI accelerates the speed and scale of cyberattacks, defenders are being asked to secure increasingly complex digital environments with approaches built for a different era,” the company said. “Security teams are often forced to piece together signals, context, and risk insights across vast amounts of data, making it harder to keep pace with emerging threats.” 微软表示，这些新工具旨在应对组织在防御灾难性黑客攻击方面发生的巨大转变。“随着 AI 加速了网络攻击的速度和规模，防御者被要求用针对不同时代构建的方法来保护日益复杂的数字环境，”该公司表示。“安全团队往往被迫在海量数据中拼凑信号、背景和风险洞察，这使得他们更难跟上新兴威胁的步伐。”
With last week’s OpenAI incident evoking troubling scenes straight out of the most dystopian sci-fi novels, the tools, which are currently in preview mode, deserve a healthy dose of caution that Microsoft made no mention of. They should be closely scrutinized and evaluated before being used in production. On the other hand, there are clear risks for not adopting such tools. Balancing the risks of using AI agents versus the threat of avoiding them is a work in progress with no clear answers for now. 鉴于上周 OpenAI 的事件让人联想到反乌托邦科幻小说中令人不安的情节，这些目前处于预览模式的工具值得保持高度警惕，而微软对此只字未提。在投入生产环境使用之前，它们应当经过严格的审查和评估。另一方面，不采用此类工具也存在明显的风险。如何在利用 AI 代理的风险与规避它们的威胁之间取得平衡，目前仍是一个正在进行中的课题，尚无明确答案。