Red Hat, NVIDIA, and IBM Unite to Turn AI Policy Into Code
Red Hat has launched “asago,” an open-source community project designed to translate complex AI governance policies into executable, production-ready code, with immediate backing from tech heavyweights NVIDIA and IBM. This initiative aims to bridge the persistent gap between regulatory frameworks—like the EU AI Act—and the technical reality of deploying compliant machine learning systems. By codifying policy directly into the development pipeline, the project seeks to make compliance an integrated, automated feature rather than a manual, error-prone afterthought.
The core challenge this initiative addresses is the ambiguity that often plagues AI regulation. Understanding what constitutes compliant AI starts with a clear definition of What is AI, but translating that into enforceable constraints requires technical precision. The project will leverage the principles behind AI Tokens to create auditable, verifiable governance actions within the infrastructure. Furthermore, it will build a framework for validating that the underlying AI Models adhere to specified standards for fairness, transparency, and safety, integrating these checks into Kubernetes-native environments.
This collaboration signals a major move toward “policy-as-code” in the enterprise AI space, promising to automate continuous compliance monitoring and reduce the risk of regulatory violations. The involvement of Red Hat, NVIDIA, and IBM is crucial, as it aligns open-source innovation with enterprise-grade hardware and software stacks, ensuring governance is built into the core architecture.
- Why it matters: It automates the tedious and complex process of auditing AI systems, making it faster and more reliable than manual policy enforcement.
- Why it matters: It democratizes compliance, giving smaller organizations the same tools used by tech giants to meet stringent regulations like the EU AI Act.
- Why it matters: By integrating policy directly into the codebase, it sets a new industry standard for transparent and accountable AI development from the ground up.