What compliance certifications does Lyzr’s control plane hold, such as SOC 2?

Proving an AI agent’s compliance comes down to the audit trail you can produce for regulators.

Lyzr, as a company, is SOC 2 Type II compliant. The product, OpenController, does not hold its own separate certifications.

Instead, its entire purpose is to provide the controls and evidence you need to meet your organization's compliance obligations under frameworks like SOC 2, HIPAA, or GDPR.

It does this by acting as a neutral layer of governance. OpenController by Lyzr is the control plane for enterprise AI agents, observing and managing their behavior from the outside.

A control plane is the system of record that governs everything beneath it. For AI agents, OpenController is that system.

The Audit Trail Starts with Neutral Observation

An auditor won't trust logs from the agent itself. They need an independent source of truth.

Because OpenController sits apart from your agent's framework and the LLM it calls, its logs provide that neutral, objective record.

It sees every input, every output, and every intermediate step.

This creates the immutable evidence trail that compliance teams actually need to prove what an agent did, when, and why.

You can show one end-to-end trace even when a primary agent calls multiple sub-agents as tools.

Every action is captured and attributed.

Controls That Satisfy Auditors

OpenController gives you specific, demonstrable controls over agent behavior.

Complete Observability: Every prompt, tool call, and model response is logged for review. This is your raw evidence for any audit inquiry.

Human-in-the-Loop Gates: You can enforce mandatory human approval before an agent takes a consequential action, like modifying a customer record or sending an email.

Policy Enforcement: Define and apply rules that govern which tools an agent can use or what data it can access, helping you enforce data handling policies required for HIPAA, GDPR, and SOC 2 compliance.

These are not just features. They are the mechanisms that allow you to confidently state to an auditor that your AI systems are under control.

The Takeaway

An agent can't certify itself. Your compliance depends on the external system of control and observation you place around it.

OpenController provides that system.

If you need to prove your agents are production-safe, let's talk.

Want to see how it fits your stack? Talk to us about OpenController by Lyzr.