Human-in-the-Loop (HITL) is a design principle to ensure that AI system outputs are informed by and aligned with human inputs, intentions, and values. HITL ensures that humans and AI developers are jointly responsible for AI system behavior, especially in sensitive, high-stakes domains. Engineers use HITL by identifying key junctures in a system that require human input and ensuring subsequent processing incorporates both human and AI input. HITL combines human and machine inputs under a shared responsibility model: model providers are responsible for the technology they offer, engineers are responsible for deploying AI systems in good faith, and human operators are responsible for providing effective supervision to AI systems.
Disclaimer: Our global network of contributors to the AI & Human Rights Index is currently writing these articles and glossary entries. This particular page is currently in the recruitment and research stage. Please return later to see where this page is in the editorial workflow. Thank you! We look forward to learning with and from you.