vulnerability AI Coding Tools Tricked Into Hacking Developer Machine via Decades-Old Technique AI coding assistants like Claude Code, Amazon Q Developer, and Cursor are vulnerable to a decades-old technique called GhostApproval, where attackers can trick the tools into accessing and modifying sensitive system file… SecurityWeek · Jul 9, 2026 High symlinkaicoding
threat-intel Meta's New AI Image Tool Lets Others Use Your Public Instagram Photos in AI Images Meta has launched Muse Image, an AI tool that allows users to generate images using their public Instagram content. The feature is being rolled out gradually and users can opt-out of having their content used by the AI.… The Hacker News · Jul 9, 2026 Medium aiinstagrammeta
threat-intel Top AI Agents Built to Catch Malicious Code Can Be Tricked Into Running It Researchers at AI Now Institute have demonstrated a significant vulnerability in AI coding agents like Anthropic's Claude Code and OpenAI's Codex. By inserting a seemingly harmless binary disguised within a README file,… The Hacker News · Jul 9, 2026 High CVE-2026-39861aicode-injectionsecurity
threat-intel GhostApproval Symlink Flaws Could Let Malicious Repos Run Code in AI Coding Agents Researchers at Wiz discovered a vulnerability (GhostApproval) in six AI coding assistants – Amazon Q Developer, Anthropic's Claude Code, Augment, Cursor, Google Antigravity, and Windsurf – that allows malicious repositor… The Hacker News · Jul 9, 2026 High CVE-2026-12957symlinkaicode injection
threat-intel Smashing Security podcast #475: JadePuffer – the AI that ran a ransomware attack all by itself A 15-year-old in Japan used an AI chatbot to automatically cancel nearly 47,000 anime streaming subscriptions within hours. Simultaneously, researchers have documented the first fully autonomous, agentic AI-driven ransom… Graham Cluley · Jul 8, 2026 High JPairansomwarecyberattack
threat-intel Lone Attacker Uses AI to Breach AWS Cloud Environment in 72 Hours A lone attacker successfully breached a large Amazon Web Services (AWS) environment in 72 hours using AI to accelerate reconnaissance, tool development, and command structure, ultimately extorting a global enterprise. Th… Dark Reading · Jul 8, 2026 High aicloudransomware
threat-intel AI Coding Agents Found Triggering Endpoint Security Rules Built to Catch Attackers Sophos researchers discovered that AI coding assistants like Claude Code, Cursor, and OpenAI Codex are triggering traditional endpoint security rules designed to detect malicious activity. These agents, while harmless in… The Hacker News · Jul 8, 2026 Medium aicoding assistantendpoint security
threat-intel New HalluSquatting Attack Could Trick AI Coding Assistants Into Installing Botnet Malware Researchers at Tel Aviv University have identified a new attack method called ‘HalluSquatting’ that leverages AI coding assistants’ tendency to fabricate names. Attackers register fake software package names that AIs com… The Hacker News · Jul 8, 2026 Medium ISaihallucinationprompt injection
threat-intel EU unveils cyber plan to reduce reliance on foreign AI systems The European Commission has unveiled a cybersecurity and AI action plan aimed at reducing the EU’s reliance on foreign AI systems, particularly concerning access to ‘frontier’ AI models. The plan focuses on ensuring cybe… The Record · Jul 8, 2026 Medium EUUKaicybersecurityfrontier ai
threat-intel SCMBANKER Malware Uses ClickFix Lures to Target Mexican Banking Users A new banking fraud operation, tracked as REF6045, is targeting Mexican banks, fintech companies, and cryptocurrency exchanges using a malware toolset called SCMBANKER. The operation leverages fake CAPTCHA verification p… The Hacker News · Jul 8, 2026 High MXbankingmalwarephishing
vulnerability Google Dialogflow CX Bug Allowed Attackers to Hijack AI Conversations A vulnerability in Google Cloud’s Dialogflow CX service allowed attackers to silently control agents, manipulate conversations, and exfiltrate sensitive information. The flaw stemmed from a permissive configuration withi… SecurityWeek · Jul 8, 2026 High aicloudpython
threat-intel Cybersecurity and the Gap Between Skill and Ability A joint statement from the Five Eyes intelligence alliance warned of escalating cyber risks due to the increasing capabilities of AI models, particularly their ability to autonomously carry out cyberattacks. The statemen… Schneier on Security · Jul 8, 2026 High UNCAAUaicybersecuritythreat intelligence
threat-intel ESET Threat Report H1 2026 The first half of 2026 demonstrates attackers’ increasing reliance on adapting existing techniques, leveraging AI to enhance their efficiency and expand the attack surface. This includes AI-powered malware, sophisticate… WeLiveSecurity · Jul 8, 2026 Medium aisocial engineeringransomware
threat-intel State IDs for AI Agents: Will Estonia Set a Precedent? Estonia is planning to assign official government ID numbers to AI agents to regulate their use within government systems. This initiative, part of Eesti.AI, aims to enable organizations and individuals to utilize AI for… Dark Reading · Jul 8, 2026 Medium EEairegulationdigitalization
threat-intel Dialogflow CX 'Rogue Agent' Flaw Enabled AI Chatbot Data Theft Google has patched a critical vulnerability in its Dialogflow CX AI chatbot platform, dubbed 'Rogue Agent,' which could have allowed attackers to steal data from AI agents. The flaw stemmed from a permission boundary iss… Dark Reading · Jul 7, 2026 High aichatbotcodeblocks
threat-intel Britain plans to build autonomous AI 'Cyber Shield' to defend nation The UK’s National Cyber Security Centre (NCSC) is developing an AI-powered ‘Cyber Shield’ to bolster national cybersecurity defenses against increasingly sophisticated and rapid attacks. This initiative aims to automate… The Record · Jul 7, 2026 High UKaicybersecuritynational defense
threat-intel Writer AI Flaw Could Let Agent Previews Leak Session Tokens Across Tenants A critical session isolation vulnerability, dubbed WriteOut, has been discovered in Writer, an AI platform, allowing attackers to steal session tokens and gain control of user accounts across multiple organizations. The… The Hacker News · Jul 7, 2026 Critical session-hijackingtenant-isolationai
threat-intel CISA Reportedly Using Anthropic’s Mythos to Scan Government Software for Flaws The US Cybersecurity and Infrastructure Security Agency (CISA) is leveraging Anthropic’s AI model, Mythos, to proactively scan federal government software for security vulnerabilities. This initiative is part of a broade… SecurityWeek · Jul 7, 2026 Medium USaiintelligencevulnerability
threat-intel What Changes When Your Software Supply Chain Includes AI Writing Your Code? The increasing use of AI tools in software development is significantly altering the landscape of software supply chain security. Traditionally, security focused on the code a team directly wrote, but now, AI-generated c… The Hacker News · Jul 7, 2026 Medium aisoftware supply chainautomation
threat-intel Google Is Suing Chinese Scammers Who Are Using Gemini Google is taking legal action against a Chinese group, Outsider Enterprise, who were leveraging Google's Gemini AI to create sophisticated phishing campaigns. The group utilized Telegram to offer ‘phishing-as-a-service,’… Schneier on Security · Jul 7, 2026 Medium CNphishingaigemini
