Agentic AI identity and guardrails: what Mercari’s approach signals

TL;DR: Mercari says rapid LLM uptake pushed AI security out of functional silos and into a dedicated governance model, with attention on authentication, authorization, auditing, tool control, and least agency as agentic AI expands, according to Opal Security. The real issue is that autonomous behaviour turns identity from a static entitlement problem into a runtime governance problem, where review cycles alone cannot keep pace.

NHIMG editorial — based on content published by Opal Security: Customer Voices on how Mercari’s security team is building guardrails for the AI era

Questions worth separating out

Q: How should security teams govern AI agents that can choose tools at runtime?

A: Security teams should govern runtime authority, not just credential issuance.

Q: Why do AI agents complicate traditional IAM and NHI controls?

A: AI agents complicate IAM and NHI controls because they act in ways that are not fully predictable at provisioning time.

Q: What do organisations get wrong about authorization for agentic AI?

A: They often treat authorization as a binary check at the front door, when agentic AI needs continuous evaluation across the whole action path.

Practitioner guidance

• Define agent decision boundaries Document which actions an AI system may initiate, which tools it may chain, and where human approval is mandatory before execution continues.
• Replace shared API keys with identity-bound access Move AI workloads away from long-lived shared secrets and toward centrally managed identity that is tied to workload or service context.
• Instrument end-to-end audit trails for AI tool use Capture the model request, tool call, data accessed, and approval context so investigators can reconstruct what the agent did after the session ends.

What's in the full article

Opal Security's full article covers the operational detail this post intentionally leaves for the source:

• How Mercari structured its dedicated AI Security function across platform security, privacy, and governance teams
• The team’s internal review process for agent-builder, AI automation, and MCP server consultations
• How Mercari applies centralized gateways to LLM usage and connected tools in practice
• The specific guardrails the security team is using to make secure AI the easiest path for internal stakeholders

👉 Read Opal Security's customer voice interview on Mercari's AI security guardrails →

Agentic AI identity and guardrails: what Mercari’s approach signals?

Explore further

View Full Forum →  |  NHI Foundation Course →