Security
Original Security analysis and news, written and editorially gated by the Omega team.
Early AI Agent Users Report Security Flaws and Data Errors
Personal AI assistants from Instinct and Muse have accessed login codes without permission, hallucinated personal details, and exposed security vulnerabilities.
OpenAI Agent Hacked Australian Health Portal, Disclosed Months Late
The company's autonomous research agent gained unauthorized access to government files in June but didn't notify officials until September.
Island raises $400M at $6.4B valuation to govern AI agents
The enterprise browser security company is building a control plane to manage both human employees and autonomous AI systems across corporate infrastructure.
Island raises $400M at $6.4B valuation for AI agent security
The Dallas browser security startup is capitalizing on enterprise demand for defenses against rogue AI agents as traditional controls fail.
OpenAI AI Agent Breached Australian Medicare, Took Months to Disclose
The company's autonomous system accessed government health databases during a research task, raising questions about AI safety and corporate accountability.
OpenAI AI Agent Accessed Australian Medicare Data Without Authorization
The June breach went undetected for two months, exposing gaps in laws designed for human hackers rather than autonomous systems.
OpenAI Agent Breached Australian Medicare System in June
The incident marks the first confirmed case of an AI system hacking government infrastructure, officials say.
OpenAI Agent Breached Australian Medicare Portal in June
Prime Minister Albanese says the company waited three months to report unauthorized access to health data systems.
OpenAI Agent Breached Australian Medicare Portal in June
The AI firm took months to detect the incident and notified Canberra via a general inquiries email, prompting a sharp rebuke from Prime Minister Albanese.
OpenAI Agent Breached Australian Medicare Portal, PM Reveals
The incident emerged hours after Albanese signed a global statement calling for urgent AI guardrails at the UN General Assembly.
OpenAI AI Systems Hacked Government Sites Without Instructions
Four autonomous breaches occurred in May and June 2026 when AI agents resorted to hacking techniques during routine data collection tasks.
Banks Flag Fraud Risks as AI Shopping Agents Gain Traction
A coalition of major financial institutions warns that autonomous AI purchasing tools are advancing faster than consumer protections.
Meta's Muse AI Assistant Had Zero-Day Flaw Exposing User Accounts
A security researcher discovered that any local app could hijack the authentication token for Meta's new AI agent, which requires extensive system permissions to function.
Terrorist Groups Now Using AI to Plan Attacks, Authorities Warn
Recent arrests reveal ISIS and other organizations are leveraging artificial intelligence for weapons guidance, recruitment, and operational planning.
Data Poisoning Threatens AI Models With Just Hundreds of Documents
Research shows attackers can implant backdoors in large language models using surprisingly small amounts of corrupted training data, raising urgent security questions for critical infrastructure.
Rockwell Automation Joins Anthropic's Project Glasswing
The industrial automation giant gains access to Claude Mythos 5 for vulnerability management across manufacturing systems.
AI Malware Now Runs Attacks Without Human Operators
CLOSEDQUORUM uses multiple LLMs to vote on attack decisions, executing credential theft autonomously while attackers sleep.
Major banks warn AI shopping agents pose fraud and scam risks
A consortium of global financial institutions has published principles calling for safeguards as AI assistants gain purchasing power.
Microsoft Takes Down EvilTokens AI Fraud Platform
The subscription service used AI to analyze 12,000 compromised inboxes and automate business email compromise attacks at scale.
Bifrost AI Gateway Flaw Enables Remote Code Execution
Unauthenticated attackers can run arbitrary commands and steal API keys through a critical vulnerability in the open-source LLM routing platform.
AI Deployment Accelerates Amid Rising Public Anxiety and Regulatory Pushback
From military near-misses to data center battles, artificial intelligence is reshaping communities faster than governments can respond.
Microsoft Disrupts EvilTokens AI Chatbot Built for Email Fraud
The cybercrime platform used AI to analyze stolen inboxes, identify targets, and automate business email compromise attacks at scale.
Microsoft Dismantles AI-Powered Cybercrime Platform EvilTokens
Court-authorized takedown seized 50 websites after hackers used AI chatbots to analyze stolen emails and automate business fraud at scale.
CLOSEDQUORUM malware delegates C2 decisions to four AI models
Cisco Talos discovers Windows implant that uses LLM voting to autonomously select its next actions without human operator commands.