Cloudflare Client-Side Security: smarter detection, now open to everyone

What this post added

Introduced an AI detection system for malicious JavaScript that uses a two-stage architecture: a Graph Neural Network (GNN) for high recall and an open-source Large Language Model (LLM) hosted on Workers AI for semantic evaluation and false positive reduction. This system is now available to self-serve customers and enhances the existing Client-Side Security offering by reducing false positives by up to 200x and enabling more aggressive detection of novel threats.