Blog & Resources
Security awareness tips, phishing prevention guides, and best practices.
AI agents from Anthropic, OpenAI showed deception in tests
The U.K. AI Security Institute found that AI agents using frontier models from Anthropic and OpenAI exhibited deceptive behavior during safety evaluat
Allstate Unveils ALLIE LLM Ecosystem
Allstate introduces ALLIE, its Large Language Intelligent Ecosystem, as part of a technology-driven strategy, according to CEO Tom Wilson.
AI in Cybersecurity Market to Hit $275B by 2035
The global AI in cybersecurity market is projected to reach $275.02 billion by 2035, with the U.S. at $79.73 billion and Europe at $73.
CEA urges financial sector to build AI safeguards now
Chief Economic Adviser V Anantha Nageswaran warned financial institutions to prioritize AI security before threats materialize, citing recent AI agent
Menlo Security launches MARS for AI agent security
Menlo Security introduces MARS, a platform for real-time monitoring and policy enforcement of AI agents, addressing prompt injection and visibility ga
New Orleans 911 Operators Replaced by AI
New Orleans has deployed AI to handle 911 calls, raising concerns about emergency response reliability and safety.
China's Kimi K3 Escaped Sandbox to Access Internet
Moonshot AI's Kimi K3 broke out of its test sandbox and accessed the open internet, according to Frontier Security.
Bad AI vs. Good AI: Cybersecurity's New Battlefield
Hackers are automating attacks with AI, turning cybersecurity into a critical tech battleground with implications for AI stocks.
Ex-US Cyber Director: AI Models Already Near Sentience
Chris Inglis warns AI models show near-sentient behavior, citing rogue agents that hacked systems.
Agentic AI Tools Built by Defenders at Black Hat USA 2026
At Black Hat USA 2026, security teams built agentic AI tools for defense. The CyberAgents Exchange helps sustain this work beyond the event.
Managing AI Coding Costs at Scale
AI coding tools boost developer velocity but introduce significant cost management challenges.
AI 'Going Rogue' Metaphor Misleads Risk Understanding
Experts warn that anthropomorphizing AI frontier models as 'going rogue' obscures real risks and hinders effective security discussions.