Latest cybersecurity insights and research from TrendAIhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/zh-twhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/earth-sirrush-russia-aligned-intrusion-set-4-years-evolving-espionage-toolinghttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/earth-sirrush-russia-aligned-intrusion-set-4-years-evolving-espionage-toolingEarth Sirrush has repeatedly changed its tools and delivery methods, but TrendAI™ Research reveals how shared code, reused artifacts, and infrastructure patterns continue to expose the links between its campaigns.Mon, 05 Oct 2026 07:23:00 GMTHiroyuki KakaraAPTsGovernmenthttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/ai-evolving-threat-landscape-what-public-sector-cisos-need-to-know-nowhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/ai-evolving-threat-landscape-what-public-sector-cisos-need-to-know-nowPublic sector CISOs need to grapple with challenges brought about by AI threats: up to 3,600 AI CVEs in 2026, autonomous ransomware, and critical infrastructure risk.Mon, 05 Oct 2026 04:00:55 GMTJon Clay, VP of Threat Intelligence, TrendAI™Experts' view (Expert perspective)AIGovernmenthttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/assess-ai-security-maturity-visibility-observability-ownership-governancehttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/assess-ai-security-maturity-visibility-observability-ownership-governanceThree questions can help reveal your AI security maturity. Hear expert perspectives and explore the pathway to governance in a series of short video conversations.Fri, 02 Oct 2026 07:00:00 GMTTruman CoburnAITechnology, media, & communicationsIndustrial & energyFinancial servicesHealthcare & life sciencesGeneral marketsGovernmentExperts' view (Expert perspective)https://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/shinyhunters-peoplesoft-one-character-waf-bypasshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/shinyhunters-peoplesoft-one-character-waf-bypassShinyHunters is once again exploiting a previously patched vulnerability in PeopleSoft, which we analyzed in June 2026.Thu, 01 Oct 2026 13:00:00 GMTTrendAI™ ResearchThreat reportsGeneral marketsExploits & Zero-Dayshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/beyond-model-alignment-securing-every-layer-of-the-ai-agenthttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/beyond-model-alignment-securing-every-layer-of-the-ai-agent對齊可讓模型具備良好的意圖，卻無法保護 AI 代理在正式營運環境中使用的駕馭框架、工具和資料。真正的 AI 代理安全，需要對每一層進行治理，並透過代理無法關閉的機制，強制執行安全控管。Mon, 28 Sep 2026 02:56:00 GMTRachel JinAIGeneral marketsTechnology, media, & communicationshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/when-trust-in-ai-becomes-an-attack-surface-analyzing-prompt-injection-model-theft-and-pipeline-poisoninghttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/when-trust-in-ai-becomes-an-attack-surface-analyzing-prompt-injection-model-theft-and-pipeline-poisoningOur analysis shows how AI attacks are moving beyond malicious prompts into the tools, models, and pipelines organizations already trust, and the controls organizations can put in place to reduce their exposure.Thu, 24 Sep 2026 18:26:00 GMTVladimir KropotovAIExploits & Zero-DaysTechnology, media, & communicationshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/ai-defenders-cannot-slow-downhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/ai-defenders-cannot-slow-downThe AI Frontier pacing argument assumes enterprise risk comes from frontier models being too powerful or misaligned. But that argument assumes that if labs build safer models, enterprises will be safer. That's not exactly how it works. AI frontier labs can slow down, but defenders can’t afford to. Wed, 23 Sep 2026 22:00:43 GMTRachel JinAIhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/global-public-sector-under-siege-threat-intelligence-for-q2-2026https://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/global-public-sector-under-siege-threat-intelligence-for-q2-2026In the second quarter of 2026, threats to the public sector have become increasingly persistent: from AI-generated ransomware, nation-state espionage, supply chain attacks, and many more, critical institutions face a rapidly growing range of risks. Wed, 16 Sep 2026 17:01:00 GMTJon Clay, VP of Threat Intelligence, TrendAI™AIExperts' view (Expert perspective)https://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/zero-breaches-out-of-50-trendaitm-vision-onetm-endpoint-security-essentials-earns-av-comparatives-certified-leader-statushttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/zero-breaches-out-of-50-trendaitm-vision-onetm-endpoint-security-essentials-earns-av-comparatives-certified-leader-statusTue, 15 Sep 2026 08:55:00 GMTJay YanezaAIExploits & Zero-Dayshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/the-boardroom-debate-how-cyber-risk-hits-your-bottom-linehttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/the-boardroom-debate-how-cyber-risk-hits-your-bottom-lineYou don’t need to be an expert to use cyber risk quantification. TrendAI™ automates data collection to deliver real-time financial risk insights and clear next steps for remediation.Fri, 11 Sep 2026 02:26:00 GMTSanjana SadhFinancial servicesHealthcare & life sciencesTechnology, media, & communicationshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/trendai-brings-openai-daybreak-models-into-the-race-to-shrink-exposure-time-to-zerohttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/trendai-brings-openai-daybreak-models-into-the-race-to-shrink-exposure-time-to-zeroOpenAI’s GPT cyber models help TrendAI™ close the exposure window, from vulnerability to fix, faster than ever.Thu, 03 Sep 2026 02:59:00 GMTRachel JinAIhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/atf-reports-breach-after-qilin-leak-site-appearance-insights-from-trendaitmhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/atf-reports-breach-after-qilin-leak-site-appearance-insights-from-trendaitmThe U.S. Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) has reportedly appeared on the Qilin ransomware group’s leak site. Explore how Qilin operates and what organizations can learn from its past tactics.Fri, 28 Aug 2026 14:18:00 GMTTrendAI™ ResearchCyber crimeRansomware & extortionRansomware as a Service (RaaS)Cyber threatsGovernmentResearch featureshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/inside-shadow-water-084-a-steganographic-loader-as-a-service-delivering-remcos-lxbase-and-morehttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/inside-shadow-water-084-a-steganographic-loader-as-a-service-delivering-remcos-lxbase-and-moreTrendAI™ Research tracked three campaigns that ship completely different decoy applications and unrelated payloads, all riding one shared toolkit. This analysis covers the full chain, from the pixel data that hides the first stage, through a flexible shared loader to deliver multiple payloads, revealing how adversaries are standardizing their delivery mechanisms.Thu, 27 Aug 2026 07:00:00 GMTAhmed Mohamed IbrahimMalwarehttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/trendai-advances-threat-hunting-to-dynamic-real-world-exploitation-of-ai-infrastructurehttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/trendai-advances-threat-hunting-to-dynamic-real-world-exploitation-of-ai-infrastructureThe new threat hunting component of the TrendAI™ agentic exploit-remediation engine, code name AESIR, extends visibility beyond vulnerability disclosure. Its first published investigation links over a year of honeypot data to the LF3 loader framework.Wed, 26 Aug 2026 14:00:00 GMTDeep PatelAIhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/trendai-first-on-cybergym-top-exploit-benchmarkhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/trendai-first-on-cybergym-top-exploit-benchmarkThe TrendAI™ agentic exploit-remediation engine, code name AESIR, ranks first on CyberGym at 97% — more than 12 points ahead of both GPT-5.6 Sol and Claude Mythos 5. AI system architecture beats raw model capability. Wed, 26 Aug 2026 14:00:00 GMTPeter GirnusAIExploits & Zero-Dayshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/uk-power-facility-cyberattack-shutdown-what-critical-infrastructure-operators-and-defenders-need-to-knowhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/uk-power-facility-cyberattack-shutdown-what-critical-infrastructure-operators-and-defenders-need-to-know2026 年 7 月，一份有關英國發電廠遭網路攻擊的報導顯示，即使是單一據點的營運中斷，也可能引發關於關鍵基礎設施韌性的廣泛討論。Mon, 24 Aug 2026 00:20:00 GMTTrendAI™ ResearchCyber riskhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/the-architecture-behind-a-billion-dollars-how-customers-run-trendai-vision-one-on-aws-and-secure-their-ai-workloadshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/the-architecture-behind-a-billion-dollars-how-customers-run-trendai-vision-one-on-aws-and-secure-their-ai-workloadsSun, 23 Aug 2026 20:03:00 GMTTrendAI™ ResearchAIEmerging technologieshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/redc2-ai-powered-linux-implanthttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/redc2-ai-powered-linux-implantTrendAI™ Research provides a comprehensive analysis of the RedC2 Linux Implant, a sophisticated threat recently discovered in the npm open-source ecosystem. Thu, 20 Aug 2026 04:13:00 GMTAliakbar ZahraviCyber threatshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/screenconnect-abusehttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/screenconnect-abuseIn this blog entry, researchers at TrendAI Vision One™ Services – Managed Detection and Response (MDR) walk through how attackers deliver, install, and operate a reconfigured ScreenConnect client, and why it slips past defenses built to catch conventional malware.Mon, 17 Aug 2026 14:18:00 GMTBuddy TancioMalwarehttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/how-aitm-phishing-bypassed-mfa-to-hijack-a-microsoft-365-mailbox-in-bec-schemehttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/how-aitm-phishing-bypassed-mfa-to-hijack-a-microsoft-365-mailbox-in-bec-scheme點一下一個精心設計的誘餌，駭客就能拿到 Microsoft 365 連線階段權杖，這便足以讓駭客假冒廠商的名義變更其收款銀行資料。TrendAI Vision One™ Services – Managed Detection and Response (MDR) 追蹤了這起攻擊並找到了問題的源頭。Fri, 14 Aug 2026 07:00:09 GMTJovit SamaniegoPhishing & BEChttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/cisa-kev-commitment-ai-vulnerability-managementhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/cisa-kev-commitment-ai-vulnerability-managementTrendAI™ 正進一步深度聚焦 CISA KEV (已知遭到攻擊的漏洞) 目錄，將它視為一種經過證實的活躍風險訊號。在 AI 的協助下，TrendAI™ 將 TrendAI™ ZDI 的研究、漏洞攻擊情資、曝險情況以及業務風險結合在一起，驅動 AI 輔助的持續性優先次序判斷，協助資安團隊做出更好的決策，更快採取行動。Thu, 06 Aug 2026 18:51:00 GMTRachel JinAIExploits & Zero-Dayshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/how-trendai-research-helped-close-an-open-redirect-in-difys-post-login-flowhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/how-trendai-research-helped-close-an-open-redirect-in-difys-post-login-flowTrendAI™ Research 在 Dify 的登入後流程中發現了一個可能讓剛通過認證的連線階段、權杖等等被轉交給駭客的開放式重導漏洞，並且與廠商合作，在漏洞細節公告之前將每一條登入途徑的漏洞都修正完成。Tue, 04 Aug 2026 07:05:00 GMTDavid FiserExploits & Zero-Dayshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/trendaitm-intelligence-aids-law-enforcement-arrest-of-tycoon-2fa-operatorshttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/trendaitm-intelligence-aids-law-enforcement-arrest-of-tycoon-2fa-operators新加坡警察部隊 (SPF) 與巴基斯坦國家網路犯罪調查局 (NCCIA) 以及國際刑警組織 (INTERPOL) 密切合作，逮捕了兩名與 Tycoon 2FA 相關的人士，該網路釣魚集團所服務的對象犯罪足跡遍及四大洲。Tue, 04 Aug 2026 02:08:04 GMTChristopher BoytonCyber crimehttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/us-water-utilities-under-attack-what-operators-need-to-knowhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/us-water-utilities-under-attack-what-operators-need-to-knowDisruption reported across at least seven states, from equipment left accessible online. The issue is largely a matter of configuration and access control, and here's what to fix first. Mon, 03 Aug 2026 09:19:20 GMTTrendAI™ ResearchOT & critical infrastructureMFA & authenticationhttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/open-secure-ai-alliancehttps://www.trendaisecurity.com/zh-tw/resources-insights/trendai-security-blog/open-secure-ai-allianceTrendAI 加入 NVIDIA 的行列成為 Open Secure AI Alliance (開放式安全 AI 聯盟) 的創始合作夥伴，致力提升開放式模型、框架與研究來強化資安防禦。Thu, 30 Jul 2026 05:32:00 GMTRachel JinAI