Nuggets Unveils Authority Control Plane for Secure AI Deployment

London-based identity and trust technology provider, Nuggets, has launched the Authority Control Plane (ACP), a governance layer designed to manage autonomous AI agents interacting with enterprise systems. This product aims to address a critical gap in existing security infrastructure by verifying not just who an agent is, but also whether each action it takes is authorized in real-time.

The ACP evaluates multiple factors before allowing any agent action: verified identity, delegated authority, organizational policies, and runtime context. Administrators can define granular controls including system access permissions, spending limits, approval workflows, and time-based expiry for delegated authorities. The company emphasizes a “failing closed” approach with an estate-wide kill switch that instantly suspends all agent authority.

Key Features of the Authority Control Plane:

  • Real-time authorization enforcement: Blocks unauthorized actions at execution rather than flagging them post-fact in logs\n* Cryptographic Action Receipts: Create tamper-proof records showing who acted, under what authority, and why each action was permitted or denied\n* Granular policy controls: Define precise scopes of agent access including resource limits, approval triggers, and time constraints\n* Open architecture: Integrates with existing identity infrastructure (IAM/PAM) rather than replacing it\n* Kill switch functionality: Immediate suspension of all agent authority across the entire organization

The launch arrives amid increasing regulatory scrutiny of AI systems. With frameworks like the EU’s AI Act and financial services regulations emphasizing human oversight and auditability, Nuggets positions the ACP as a solution to help organizations meet compliance requirements while safely deploying autonomous agents.

“Customers told us they can’t move agents to production because they can’t prove what an agent was allowed to do,” explained Seema Khinda Johnson, co-founder and chief commercial officer of Nuggets. “That’s why we built the Authority Control Plane: to set what agents can do, enforce it as they act, and prove it to anyone who asks.”

The company has released an open-source integration with LangChain demonstrating practical enforcement behavior and plans to expand support for other agent frameworks. Pricing details and early customer deployments were not disclosed in the announcement.