“LaunchDarkly helps teams safely manage code and AI agents in production with feature flags, progressive delivery, automated rollback, and runtime control.”
“AI agent governance controls what AI agents are allowed to do in production, who can change their behavior, and how those changes are reviewed and recorded. This means access control, approval workflows, policy enforcement at runtime, and an audit trail.”
“AI governance is usually organizational: model risk policy, review boards, documentation, and standards. Agent governance is operational. It covers the controls that apply to a running agent, including who can change its prompts and models, which behavior each audience gets, and how quickly you can stop it.”
“Agents act autonomously, and their behavior changes with prompts and models. A small configuration change can shift what an agent does in production, so the governance layer has to sit at runtime rather than in code review.”
“Yes. Targeting rules let you serve different agent configurations depending on geography, entitlement, customer tier, or any other context attribute. This lets agents behave differently where regulations or contracts require it.”
Download this profile and inspect its supporting evidence.