AI 網路犯罪七大運作原則:為何駭客追求規模化,而非全自動攻擊
AI 尚未催生全新的網路犯罪模式,卻正讓既有犯罪手法變得更快速、成本更低,也更容易規模化。 趨勢科技 TrendAI™ 最新研究歸納出網路犯罪者採用 AI 的七大原則,解析 AI 如何提升網路釣魚與詐騙活動的執行效率、加速被竊資料的分析與處理,並逐步應用於惡意程式開發。同時指出,AI 代理程式雖已被用於實際入侵,但其自主攻擊能力仍受到技術可靠性、基礎設施存取權及人為監督等條件限制。

AI 尚未催生全新的網路犯罪模式,卻正讓既有犯罪手法變得更快速、成本更低,也更容易規模化。 趨勢科技 TrendAI™ 最新研究歸納出網路犯罪者採用 AI 的七大原則,解析 AI 如何提升網路釣魚與詐騙活動的執行效率、加速被竊資料的分析與處理,並逐步應用於惡意程式開發。同時指出,AI 代理程式雖已被用於實際入侵,但其自主攻擊能力仍受到技術可靠性、基礎設施存取權及人為監督等條件限制。
法律事務所的資安風險,未必會隨勒索軟體攻擊結束而消失。趨勢科技 TrendAI™ 研究團隊發現,網路犯罪者會在地下市場反覆兜售法律事務所的系統存取權限與機密資料,部分交易甚至在資安事件公開前就已出現,事件結案後仍持續流通。研究顯示,公開的勒索軟體資料外洩紀錄,可能只是法律產業所面臨資安威脅的一部分。
AI 代理程式(AI Agent)不只是聊天機器人,更能自主規劃與執行任務、呼叫外部工具,甚至與其他 AI 代理程式協作。隨著自主性提高,企業也面臨提示詞注入攻擊(Prompt Injection)、權限過大及機密資訊外洩等資安風險。 趨勢科技 TrendAI™ 透過實際測試,探討如何結合 NVIDIA OpenShell 與 TrendAI Vision One™,運用最小權限原則及涵蓋預防、偵測與應變的多層安全控管機制,協助企業降低 AI 代理程式帶來的資安風險,強化安全治理。
Google Chrome 會在符合條件的桌面裝置上自動下載約 4.27 GB 的 Gemini Nano 語言模型,使瀏覽器成為裝置端 AI 執行環境。然而,在現行架構下,任何 HTTPS 網站都能在沒有使用者同意提示、使用狀態指示及速率限制的情況下,於背景呼叫該模型,衍生提示詞注入(Prompt Injection)、運算資源濫用(Parasitic Computing),以及實驗性工具呼叫功能所涉及的潛在 Return-to-Tool(RTT)攻擊風險。 本文深入解析這三類資安風險,並提供企業資安團隊與開發者具體的防護及治理建議。
As Ethereum currently sells for over $1,200 per coin, a new ransomware variant that appears to be the first to accept the currency as ransom payment surfaces.
Read ArticleSecurity researchers discovered an installer for a Monero miner intended to transmit the cryptocurrency to Kim Il Sung University (KSU), North Korea. Meanwhile, a Reddit user found a Monero miner in the BlackBerry mobile site.
Read ArticleMultiple vulnerabilities and a hard-coded backdoor were discovered in Western Digital’s My Cloud network attached storage (NAS) devices that could allow remote attackers to gain unrestricted root access.
Read ArticleSecurity researchers reported a security flaw in Google Apps Script that can enable hackers and cybercriminals to deliver Google Drive-hosted malware when abused and exploited.
Read ArticleMicrosoft is implementing a new process for delivering Windows updates and patches: Registry keys are now required. Here's what users and businesses need to know.
Read ArticleThe U.S. Department of Homeland Security (DHS) confirmed a data breach that compromised the information of 250,000 DHS individuals involved in ongoing DHS criminal investigations.
Read ArticleApparel retailer Forever 21 disclosed its findings on the data breach it reported last November 2017, revealing signs of point-of-sale (PoS) malware and unauthorized network access into affected PoS devices.
Read ArticleThe source code of the Satori internet-of-things (IoT) botnet, pegged as Mirai's successor, was posted online on Pastebin. What does this mean in the IoT threat landscape, and what can users do against it?
Read ArticleHere are some of 2017's notably abused system administration tools and protocols as well as best practices enterprises can do against threats that exploit them.
Read ArticleTrend Micro observed a recent spate of attacks designed to deliver a webminer for the Electroneum (ETN) cryptocurrency.
Read Article