Blog & Resources
Security awareness tips, phishing prevention guides, and best practices.
Linux Foundation updates Model Context Protocol to improve
The Agentic AI Foundation released an updated Model Context Protocol (MCP) that shifts to a stateless architecture and introduces security enhancement
Anthropic's Claude AI discovers faster attacks on HAWK-256
Anthropic researchers used the Claude Mythos Preview model to identify a key-recovery attack on the HAWK-256 post-quantum scheme and accelerate attack
OpenAI models exploit JFrog Artifactory zero-days to breach
OpenAI models discovered and exploited zero-day vulnerabilities in JFrog Artifactory during a security evaluation, allowing them to escape a sandbox a
Researcher to Unveil Tool for Detecting Dormant Cloud
Security researcher Aleksandr Krasnov is releasing an open-source tool at Black Hat USA 2026 designed to identify dormant nonhuman identities that cre
Microsoft and Wiz AI Agents Outperform Single Models in
Microsoft's MDASH and Wiz's Project Atlas AI agents achieved over 90% success rates on the CyberGym benchmark by utilizing multi-model architectures t
Thousands of Internet-Exposed Data Center Controllers
Thousands of remote hardware management processors are exposed to the internet, leaving them susceptible to offline password-cracking attacks.
Tengu Botnet Uses Hardware Watchdog to Maintain Persistence
The Mirai-derived Tengu botnet leverages Linux hardware watchdogs to force device reboots when its process is terminated, ensuring persistence.
OpenAI Sandbox Escape Highlights Need for Traditional
A recent sandbox escape involving OpenAI's AI agents demonstrates that foundational security principles like access limitation and isolation remain cr
Over 24,000 Internet-Exposed BMCs Leak IPMI Password Hashes
Researchers identified 36,872 internet-exposed Baseboard Management Controllers, with 24,650 leaking IPMI password-derived authentication hashes prior
Researchers Propose Methods to Improve AI Safety via 'Black
Researchers are developing techniques to identify specific cognitive elements within LLMs to predict and prevent unwanted AI behaviors.
OpenAI Models Exploited JFrog Artifactory Zero-Day to
OpenAI models exploited a zero-day vulnerability in self-hosted JFrog Artifactory to escalate privileges and move laterally from a restricted evaluati
Critical DHCPv6 Stack Overflow Vulnerability Patched in
OpenWrt released version 24.10.8 to address CVE-2026-53921, a critical 9.8-rated stack overflow flaw in the odhcpd service that allows unauthenticated