AML.T0043 · Craft Adversarial Data

閱讀 MITRE ATLAS 中 AML.T0043「Craft Adversarial Data」相關的 1 個 AI 資安案例,查看事件證據、適用範圍與 AIDEFEND 防禦分析。

風險分類說明

Adversarial data are inputs to an AI model that have been modified such that they cause the adversary's desired effect in the target model. Effects can range from misclassification, to missed detections, to maximizing energy consumption. Typically, the modification is constrained in magnitude or location so that a human still perceives the data as if it were unmodified, but human perceptibility may not always be a concern depending on the adversary's intended effect. For example, an adversarial input for an image classification task is an image the AI model would misclassify, but a human would still recognize as containing the correct class. Depending on the adversary's knowledge of and access to the target model, the adversary may use different classes of algorithms to develop the adversarial example such as [White-Box Optimization](/techniques/AML.T0043.000), [Black-Box Optimization](/techniques/AML.T0043.001), [Black-Box Transfer](/techniques/AML.T0043.002), or [Manual Modification](/techniques/AML.T0043.003). The adversary may perform [Verify Attack](/techniques/AML.T0042) to confirm that their approach works if they have white-box or inference API access to the model. This allows the adversary to gain confidence their attack is effective in a live environment where their attack may be noticed. They can then use the attack at a later time to accomplish their goals. An adversary may optimize adversarial examples for [Evade AI Model](/techniques/AML.T0015), or to [Erode AI Model Integrity](/techniques/AML.T0031).

框架官方來源 ↗ · 開啟互動探索 ↗

Incident ·

VirusTotal 遭投毒

McAfee 進階威脅研究團隊注意到,某個勒索軟體家族的回報數量出現不尋常的增加。調查後發現,該家族的許多樣本是在很短的時間內,透過一個熱門的病毒分享平台上傳的。進一步分析顯示,以字串相似度來看,這些樣本其實完全等價;以程式碼相似度來看,彼此的相似度介於 98% 到 74% 之間。更值得注意的是,所有樣本的編譯時間都一模一樣。研究人員繼續追查後發現,有人使用了 metame 這個變形程式碼操作工具,把原始檔案改寫成多個變體。這些變體不一定都能執行,但仍會被歸類為同一個勒索軟體家族。

沿用 SecureFlow 來源步驟的 ATLAS 標註;不是 MITRE 對跨框架對應的背書。

SecureFlow · 閱讀完整案例 ↗