AIShield Guardian is a middleware security layer designed to help enterprises safely adopt Generative AI (GenAI) and Large Language Model (LLM) technologies such as ChatGPT. It operates by scrutinizing both user inputs sent to LLMs and the outputs returned by them, enforcing compliance with an organization's legal, ethical, and internal policy requirements. Key capabilities include prevention of personally identifiable information (PII) leaks, intellectual property protection, and policy compliance enforcement. The product uses a role-based policy mapping system that can be customized using keywords, enabling organizations to tailor controls to different user roles within the enterprise. AIShield Guardian is implemented as a set of SDKs that can be integrated into existing workflows with approximately five lines of Python code. It is compatible with any LLM, making it applicable across a broad range of enterprise use cases. The product supports two deployment-oriented product lines as indicated by the demo form: a GenAI Guardrails variant and an ML Firewall variant, both operating under the Guardian umbrella. Use cases addressed include: - Preventing PII leakage through LLM interactions - Enforcing organizational and regulatory policy compliance - Protecting intellectual property in AI-assisted workflows - Enabling responsible and auditable experimentation with GenAI tools - Improving operational efficiency by automating policy enforcement at the AI layer