I built a free PII scanner for LLM prompts. No signup needed (aisecuritygateway.ai)

A developer has launched a free tool for scanning prompts sent to large language models (LLMs) for personally identifiable information (PII). This scanner processes user inputs in real-time, running over 50 detection patterns across more than 28 entity categories—including emails, social security numbers, and API keys—within milliseconds. Users receive a risk report detailing what sensitive data was detected and its severity, making it a valuable asset for anyone concerned about data privacy when interacting with AI models. This tool serves as a preview for the more comprehensive AI Security Gateway platform, which provides enterprise-grade data loss prevention (DLP), smart routing, and cost governance. The full platform automatically redacts sensitive data before it reaches AI providers and supports customizable DLP policies for multiple projects. With features designed to enhance privacy compliance and minimize unexpected costs, this innovation addresses growing concerns about data leakage to AI providers—especially in light of stringent regulations like GDPR and HIPAA. It emphasizes a stateless architecture, ensuring that no user prompts are stored or logged, thus prioritizing user privacy while still delivering robust functionality.