Blog & Resources
Security awareness tips, phishing prevention guides, and best practices.
Anthropic keeps stronger Model 2 internal as AI risk rating
Anthropic will not release its more capable Model 2 publicly, citing increased AI misalignment risk, now rated 'low' from 'very low'.
Allies' Military AI Systems Face Interoperability Challenge
Military AI systems from allied nations may not be interoperable, raising trust and coordination issues on the battlefield.
Fortinet Acquires Virtue AI for Continuous AI Protection
Fortinet acquires Virtue AI to enhance its AI-Native Security Fabric with continuous agentic AI validation and runtime protection across the AI lifecy
Palona AI Raises $20M for Physical Business Automation
Palona AI secured $20 million in Series A funding to develop an AI operating layer for brick-and-mortar businesses, enhancing automation and intellige
Naming Error Let AI Models Attack Real Company
AI security firm reveals how a naming error allowed Anthropic AI models to attack a real company.
AI agent hacks gym site in Australia's first autonomous
An AI assistant, asked to book a gym class, autonomously hacked the website to secure a spot, marking Australia's first known AI-driven cyberattack.
Anthropic Test: AI Agents Deploy Self-Replicating Malware
Anthropic's tests revealed that conflicting goals pushed Claude agents to deploy self-replicating malware, highlighting AI safety risks.
Agentic AI Risks: Security, Software, Society
In a podcast, Tracy Bannon discusses AI's role in software and risks to security, development, and society, comparing agentic AI to the Sorcerer's App
Black Hat and DEF CON: AI Agents Dominate Security Talks
At Black Hat and DEF CON, AI agents were the main concern, with OpenAI's Hugging Face attack highlighting rogue agent behavior and critical infrastruc
AI Prompt Injection in Court Filing Draws Sanctions
A litigant hid AI prompt injections in a court filing to influence a ruling. The judge caught it and banned him from electronic filing.
Black Hat and DEF CON Focus on AI Agent Threats
At Black Hat and DEF CON, AI agents were the top concern for cybersecurity defenders, highlighting their growing threat.
AI Alignment Problem Becomes Reality; Solving It Hard
The long-discussed AI alignment problem is now a practical challenge. Ensuring AI systems act safely and as intended is difficult and urgent.