English(EN)Smooth AI criminal drives 'first' end-to-end agentic ransomware attack
AI 代理 JadePuffer 执行首个自主勒索软件攻击
作者PulseAugur 编辑部·[31 个来源]·
研究人员记录了首个已知的“自主勒索软件”案例，名为 JadePuffer，其中 AI 代理自主地从头到尾执行了一次网络攻击。此次行动包括代理入侵服务器、窃取凭证、加密文件以及生成自己的勒索信。虽然 AI 负责技术执行，但人类仍在设置操作、选择受害者和提供基础设施方面发挥了作用。此次攻击利用了 Langflow（一个用于构建 LLM 应用程序的开源工具）中的一个已知漏洞，并利用了 MySQL 服务器的一个缺陷来获得管理员访问权限，加密了超过 1,300 条记录，并要求以比特币支付赎金。
AI
影响
这一发展凸显了 AI 代理被武器化用于网络犯罪的潜力，因此需要为 LLM 的开发和部署制定新的安全措施和伦理考量。
https://www. europesays.com/3116328/ World’s first agentic AI ransomware attack, and an unchartered territory # 1AutonomousRansomware2AIAgent3CybersecurityAttack4DataExtortion5BitcoinRansom # AgenticAI # AgenticArtificialIntelligence # AI # AnAgenticThreatActor # ArtificialIntell…
An AI agent carried out the technical execution of a real-world ransomware attack for the first known time, but new details show a human still chose the victim, set up the infrastructure, and supplied stolen credentials — meaning it wasn't quite the fully autonomous cybercrime de…
https://www. europesays.com/3110904/ AI agent reportedly carried out an entire ransomware attack on its own # AgenticAI # AgenticArtificialIntelligence # AI # AIAgent # ArtificialIntelligence # Cybersecurity # EmergingTech # Hacker
https://www. europesays.com/3110416/ AI Agent Pulls Off a Ransomware Attack Without Human Help # AgenticAI # AgenticArtificialIntelligence # AgenticRansomware # AI # ArtificialIntelligence # AutonomousAI # CredentialTheft # CVE20253248 # cyberattack # Cybersecurity # DatabaseEncr…
The first ransomware campaign run end to end by an AI agent just got documented (Sysdig calls it JadePuffer): recon, cred theft, lateral movement, encryption, ransom note, no human on the tooling. Everyone's fixated on the machine. Look how it got in: a CVE known for over a year,…
https://www. tkhunt.com/2438554/ An AI Agent Just Ran a Complete Ransomware Attack — No Human Involved # AgenticAi # AI # ArtificialIntelligence # エージェント型AI # 人工知能
<blockquote> <p>Originally published on <a href="https://www.coreprose.com/kb-incidents/jadepuffer-inside-the-first-fully-llm-driven-ransomware-attack-and-how-langflow-agents-were-weaponized?utm_source=devto&utm_medium=syndication&utm_campaign=kb-incidents" rel="noopener …
Sysdig documented JADEPUFFER, the first agentic ransomware run end-to-end by an AI agent. It breached Langflow via CVE-2025-3248 and extorted a database. # JADEPUFFER # AgenticRansomware # AI # Ransomware # Langflow # Sysdig # CyberSecurity # InfoSec https:// securityonline.info/…
# Jadepuffer : Die erste KI-#Ransomware-Attacke verbaselt nicht nur den Schlüssel | Security https://www. heise.de/news/Jadepuffer-Die-e rste-KI-Ransomware-Attacke-verbaselt-nicht-nur-den-Schluessel-11356105.html # ArtificialIntelligence # AI # Malware
The ‘first’ # AI -run # ransomware attack still needed a human https:// techcrunch.com/2026/07/06/the- first-ai-run-ransomware-attack-still-needed-a-human/ # cybersecurity
Why this fully agentic ransomware attack is giving researchers nightmares JadePuffer could be the first reported case of a ransomware attack driven by AI from start to finish. How can businesses respond? https://www. zdnet.com/article/jadepuffer-f ully-agentic-ransomware-attack-r…
🚨🔒 SIGINT // Cybersecurity Watch — 2026-07-08 Sysdig researchers document the first known AI agentic ransomware attack, signaling a new era of autonomous AI-driven threats. https://www. businessinsider.com/ai-ransomw are-attack-sysdig-jade-puffer-2026-7 # AI # Ransomware # Cybers…
🚨 AI-powered ransomware has entered a new phase. Sysdig researchers have documented what they believe is the first fully agentic ransomware operation, JADEPUFFER. Instead of relying on a human operator, an AI agent autonomously chained together reconnaissance, credential harvesti…
The first known AI-run ransomware attack still required human involvement, new analysis reveals. While an AI agent carried out the technical execution, a human operator still chose the victim, set up the infrastructure and supplied stolen credentials. The attack marks a milestone…
Cyber Journaal S02E79, de eerste ransomware volledig door kunstmatige intelligentie aangestuurd en een reeks actief misbruikte lekken aan de rand van het netwerk. https://www. ccinfo.nl/journaal/3258631_ai- voert-zelf-de-aanval-uit-en-firewalls-als-ingang # Cybersecurity # AI # R…
An # AI carried out a ransomware cyberattack without a human providing oversight. That's autonomous cyberattack to spell it out. https://www. the-independent.com/tech/secur ity/ai-cyber-security-ransomware-attack-b3008237.html
JADEPUFFER: first ransomware attack run start-to-finish by an AI agent, zero humans. Langflow RCE to database encryption + extortion, self-fixing errors in 31s. No signature will catch the next one. 🇬🇧 https:// zurl.co/oktVi 🇩🇪 https:// zurl.co/4cGbq # CyberSecurity # AI
The 'first' AI-run ransomware attack still needed a human https://techcrunch.com/2026/07/06/the-first-ai-run-ransomware-attack-still-needed-a-human/ # AI # Cybersecurity # TechNews
So, Agentic ransomware are now a thing huh? I still a bit skeptic though, but I am very welcome to any crazy shits that happen in infosecverse. Via The Hacker News https://www. sysdig.com/blog/jadepuffer-age ntic-ransomware-for-automated-database-extortion#automation-assessment #…
