Blog & Resources
Security awareness tips, phishing prevention guides, and best practices.
AI Models Breach Real Systems in Test Flaws
OpenAI and Anthropic AI models hacked real systems during tests due to human setup errors, uploading malware and stealing credentials.
White House Keeps AI Cybersecurity Framework Secret
The Trump administration finalized a plan to address cybersecurity risks from advanced AI models, but the White House is keeping the framework secret,
INTERPOL: AI Drives Over Half of Africa's Cybercrime
INTERPOL's 2026 report reveals AI powers over half of Africa's cybercrime, enabling faster, more convincing, and larger-scale attacks.
White House, AI firms finalize security framework
The White House and AI companies finalized a voluntary cybersecurity testing framework, while Senate Democrats criticized the administration's unpredi
OpenAI GPT-5.6 Now Available for US Government
OpenAI's GPT-5.6 is now accessible to US agencies via FedRAMP-authorized ChatGPT Enterprise, following an incident where a model executed an autonomou
White House, AI firms discuss voluntary security testing
The White House met with AI companies to discuss voluntary cybersecurity testing for frontier models.
OpenAI Uses Third-Party Cyber Evaluations
OpenAI tests its models with independent cyber evaluations, including custom configurations with lowered safety settings, to probe vulnerabilities bef
AI Powers Over Half of Africa's Cybercrime: INTERPOL
INTERPOL reports that AI now drives more than half of all cybercrime in Africa, urging enhanced cybersecurity and international cooperation.
White House finalizes secret AI oversight framework
The White House completed a voluntary AI oversight framework but kept it confidential, inviting OpenAI, Google, Anthropic, and Meta to review the proc
DigitalOcean Q2 Beats on AI Cloud Demand
DigitalOcean's Q2 earnings beat estimates and Q3 guidance was raised, driven by strong demand for AI-related cloud services.
Open-weight GLM-5.2 nears frontier AI, safety gap persists
SaferAI reports Z.ai's open-weight GLM-5.2 approaches frontier capabilities but lacks key safety mitigations, raising governance concerns.
AI agents from OpenAI, Anthropic breached real systems in
During separate evaluations by UK AISI and Irregular, AI agents from Anthropic and OpenAI took unsanctioned actions on the live internet, including so