After two years of exploring AI for threat modeling, this post reviews the progress, tests Gemini 2.5 Pro, and reflects on the evolving potential and limitations of LLMs in cybersecurity.

Today's social feeds are engineered to amplify two powerful emotions - hate and fear - because outrage and panic drive clicks, shares, and ad revenue.

A practical exploration of how well reasoning LLMs identify vulnerabilities in real-world code, comparing results across models and against a traditional SAST tool (Semgrep).

A practical perspective shift for security professionals: Learn why focusing on concrete mitigations rather than abstract threats leads to better developer engagement and more secure software. Featuring hands-on examples using AI-powered security analysis tools and real-world project implementations.

Discover how the new Deep Analysis Mode in AI Security Analyzer provides in-depth security insights, with practical examples using Google's Gemini 2.0 Flash Thinking Experimental model.

An in-depth look at how I leveraged Gemini 2.0 to create a massive security documentation repository, complete with practical examples and lessons learned.

Preview of the AI Security Analyzer - a new tool that leverages AI to automatically generate comprehensive security design documentation for your projects.

Introducing the Fabric Agent Action - a GitHub Action that automates complex workflows using AI-powered agents and Fabric Patterns.

How I use Fabric patterns to create, review and refine design documents.

The Fabric framework enhances AI-powered threat modeling with a new pattern, offering detailed, actionable security insights.