Topic starter
02/08/2026 11:22 am
TL;DR: AI agents create a governance gap because they can act across systems, change after deployment, and outgrow point-in-time appsec assumptions, according to Akto. Governance has to track ownership, inventory, policy, monitoring, and revalidation or the approved system drifts away from reality.
Editorial analysis by NHI Mgmt Group, based on content published by Akto: “AI Agent Security Best Practices: How to Secure Enterprise AI Agents”.
Key questions
Q: Why do AI agents create more risk than traditional automation?
A: AI agents create more risk because they can interpret context, choose actions, and invoke tools autonomously.
Q: What are the signs that AI agent lifecycle controls are failing?
A: Common signals include undocumented agents, unclear ownership, persistent tokens, broad permissions that exceed task scope, and retired agents that still have live integrations.
Q: How should security teams govern AI agents that can access enterprise systems?
A: Security teams should govern AI agents as non-human identities with explicit ownership, scoped privileges, and continuous monitoring.
Practitioner guidance
- Establish named ownership for every agent Assign business, technical, and security owners before an agent is allowed into production, and record those owners alongside the agent in the inventory so accountability survives organisational handoffs.
- Build a complete AI agent inventory Discover approved agents, internal experiments, third-party agents, MCP servers, and shadow AI across cloud, endpoints, and internal systems, then keep the inventory synced as the environment changes.
- Classify agents by risk and action scope Separate low-risk assistants from high-risk and critical agents that can touch customer data, personal data, money movement, or production changes, then require stronger controls where consequences are material.
Bottom line: AI agents create a governance gap because they can act across systems, change after deployment, and outgrow the assumptions behind point-in-time application security.
Glossary terms: AI Agent Governance, AI Agent Drift, Runtime Guardrail, Shadow AI
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →
This topic was modified 2 days ago by NHI Mgmt Group
02/08/2026 12:40 pm
Ownership is the control that turns agent governance from theory into enforcement. Without a named business owner, technical owner, and security owner, policy approval becomes symbolic and incidents stall in ambiguity. The article is right to frame accountability as a structural issue, because AI agents do not fit legacy ticket-based responsibility models. For practitioners, that means ownership must be recorded as part of the asset itself, not as an afterthought in an operations runbook.
A few things that frame the scale:
- Gartner predicts that more than 50% of successful cyberattacks against AI agents through 2029 will exploit access control weaknesses.
A question worth separating out:
Q: Should organisations prioritise runtime guardrails or agent inventory first?
A: Inventory should come first because organisations cannot enforce runtime guardrails on agents they have not discovered or assigned to owners. Once the fleet is visible, runtime guardrails become the next control layer for constraining actions in production. In practice, the two controls are sequential, not interchangeable.
👉 Read our full editorial: AI agent governance needs ownership before runtime guardrails
This post was modified 2 days ago by NHI Mgmt Group