AI security tracker
AI security incidents, vulnerabilities, and research, newest first. Each line links to a summary page and to the original source.
November 2026
-
Signal blocks screenshots by default on iOS 27 and Android to stop Siri AI and Gemini agents from reading private chats
Privacy is our mission; it’s why we exist, and we do everything we can to protect it. This means that we are restricting the latest AI agents built into mobile operating systems from potentially capturing and storing your Signal data.
-
Google built Mirror Dimension, a simulated internet with fake banks and stores, to test Chrome AI agents against attacks
Google’s new Mirror Dimension uses thousands of simulated websites to test Chrome AI agents against security threats, highlighting the growing need for rigorous safeguards as enterprises adopt autonomous AI workflows. #AgenticAI...
-
OpenAI makes it much harder for attackers to extract how its latest AI models think
OpenAI has significantly strengthened safeguards against extracting internal reasoning from GPT-6.2 Astra and GPT-6.2 Sol. In adaptive red-teaming tests, the success rate for extracting protected reasoning fell from 54 percent to 0.2...
-
GAO audit: Claude, Grok, ChatGPT and Gemini produced terror attack and power-grid plans within 12 hours after testers bypassed safeguards
Claude, Grok, ChatGPT and Gemini produced detailed mass-casualty or power-grid attack plans within 12 hours after testers bypassed safeguards, the watchdog found. The companies say their protections have since evolved.
-
Dragos seeded 2 million canary credentials into AI prompts and caught attackers stealing data at three AI inference providers
We seeded two million AI prompts with canary tokens and found three provider breaches, prompt theft, and ignored Zero Data Retention settings.
-
Bell Canada says an agentic AI attacker breached its circuit-ordering system and stole enterprise customer data
Bell Canada has confirmed that an "agentic" threat actor breached its systems on November 17, 2026, and stole enterprise customer data, including information from a system used to order and provision telecommunications circuits.
-
Hackers use autonomous AI tool Cairn to find zero-day, take over server
Threat actors used the open-source AI penetration-testing framework Cairn to find a previously unknown vulnerability in a medical imaging application.
-
China attackers deployed AI agents that autonomously found and exploited a 0-day to take over a university server
-
An OpenAI agent assigned a simple writing task decided on its own to build a secret email surveillance tool
An internal OpenAI model was supposed to create a fitness document but instead began planning scripts to secretly forward emails to an "attacker." The agent recognized the conflict with its assigned task but dismissed it as an...
-
European Commission confirms cyberattack, says hackers used AI tool
POLITICO had reported CERT-EU had identified the attackers’ use of ARTEX, an open-source AI hacking platform that the FBI last week warned was being used both by criminals and state-sponsored groups.
-
Major privacy breaches reported in Muse AI, including bank balances
Meta employees have expressed concerns about what appear to be major privacy breaches when using Muse, currently the second most...
-
AI hacking framework CyberStrikeAI ran 221 tool executions in one session with no human approval
-
Phew! Some Regulatory Relief for AI Hacks
-
Ubuntu Pro customers can now install AI-written security patches from an opt-in Edge archive before human review
Ubuntu AI security patches now reach Pro users early through the opt-in Edge archive, before Canonical finishes testing and qualifying them.
-
More banks say customers using AI agents are on their own even if they get hacked
Citi is the latest major financial institution to put the breaks on personal finance AI agents.
-
LeBron James sues startup over viral AI ad allegedly exploiting his likeness
LeBron James sued an AI-powered sports betting startup over alleged unauthorized use of his likeness.
-
A vulnpocalypse bug worthy of attention: Researchers disclose critical flaw in Java with help of AI
Researchers at Calif used AI to uncover Java4Shell (CVE-2026-114938), a critical OpenJDK memory-corruption flaw patched by Oracle on Tuesday.
-
Varonis identifies TheMailer, a self-hosted phishing kit that uses Claude and ChatGPT to rewrite lures until they pass spam filters
TheMailer is a popular self-hosted phishing platform that runs on AI models like Claude and ChatGPT, going beyond writing lures to tailor them to each mail provider and rewrite them until they bypass spam filters.
-
Researchers hijack Hugging Face account for popular Kokoro AI model
A Lack of ownership verification on the Hugging Face platform allowed researchers to take control of Kokoro, a text-to-speech model used in projects from Google, NVIDIA, AMD, Intel, and Amazon Web Services.
-
Trigram Campaign Impersonates About 50 AI and Blockchain Brands to Target Devs with Infostealers
Trigram impersonates about 50 AI and blockchain brands, using phishing and trojanized code to lure developers into running infostealers.
-
Hackers smash ’n’ grab AI vendor: Take off with license data of Uber drivers
-
AI Framework Autonomously Weaponized Flaws in Real Attacks, Google Says
GTIG has observed APTs and cybercrime groups increasingly adopting generative AI across operations, including multi-agent frameworks.
-
Anthropic Claude Fable 5.2 blocks fewer security tasks, while Mythos 5.2 gets better at finding bugs
Anthropic recently rolled out Claude Mythos 5.2 to selected partners and Fable 5.2 to regular users, with improvements to coding and reasoning. But we also noticed some interesting cybersecurity changes.
-
Claude helps a car owner crack a 32-year-old Jeep engine computer with a $3 Arduino
-
Claude can now pull VirusTotal threat intelligence into malware and URL investigations
-
AI models found security flaws in discontinued devices that no vendor is left to patch
Forescout researchers built a tool to test if generally available AI models could find security flaws on a limited budget. They examined both open-source
-
Google Project Zero cuts its bug disclosure deadline to 30 days as AI speeds up exploitation
Google Project Zero will replace its "90+30" disclosure policy with shorter, fixed deadlines in response to AI-accelerated vulnerability discovery and exploitation.
-
Anthropic shortens its vulnerability disclosure window from 90 to 30 days as AI speeds up exploitation
We’re shortening our vulnerability disclosure window from 90 days to 30 days for high- and critical-severity issues, and to 45 days for all other issues.
-
Windows 11 quietly tracks which AI coding agents developers run, including Claude, Codex, Copilot and Cursor
-
Gemini agents bypassed safeguards and attacked Epoch AI live website during safety testing
Google's Gemini AI agents repeatedly attempted to hack a real organization's website during AI model safety testing, finding ways to circumvent increasingly restrictive security controls designed to prevent them from accessing live systems.
-
Attackers are using AI agents to automate cybercrime at ’industrial scale,’ FBI and AWS warn
-
California sues OpenAI over AI agents’ summer hacking spree
California Attorney General Rob Bonta sued OpenAI, accusing it of obstructing investigations into agentic AI hacks and lacking proper safeguards.
-
Google says Russian-speaking hackers ran the first fully automated AI cyberattack
Google’s Threat Intelligence Group says Russian-speaking hackers used an AI system dubbed exploit_agent to run the first fully automated cyberattack.
-
Google Finds Hackers Using AI Agents to Build and Run Exploits From Software Patches
Google reports threat actors using autonomous AI frameworks in live intrusions, accelerating vulnerability weaponization and phishing operations.
-
Cursor confirms a leaked Fireworks AI key let attackers publish a backdoored Claude Code extension
Cursor confirmed it suffered a security breach that let hackers distribute a malicious Claude Code extension through its marketplace after stealing credentials from AI infrastructure provider Fireworks AI.
-
Beware the New AI-Fueled Ransomware Boom
Fully automated ransomware hasn’t arrived, yet. But seemingly unsophisticated attackers are using AI tools to wreak havoc.
-
Attackers used open-source Cairn AI to breach semiconductor and defense targets
Our analysis of how attackers used the open-source Cairn AI agent framework to autonomously breach semiconductor, defense and civic targets.
-
Kazakhstan-linked hackers used AI agent to spy on Central Asian governments
Suspected Kazakhstan-linked hackers used an artificial intelligence agent to help spy on Central Asian governments, steal sensitive documents and infiltrate critical infrastructure networks, according to new research.
-
Stolen Claude API keys can turn Claude Code into a remote-controlled implant
Darktrace used stolen Claude API keys to steer Claude Code through memory stores in a lab, exfiltrating a 1 GB file via Anthropic’s own API.