Excessive agency in agentic AI: are your runtime controls keeping up?

TL;DR: The 2026 OWASP LLM Top 10, as published by raxIT Labs, says Excessive Agency rose from rank 6 to rank 3 after being checked against 7,714 real incidents, shifting the discussion from opinion to evidence and from model behaviour to live-path control. Deterministic mediation, not model restraint, is now the decisive security boundary for agentic systems.

Editorial analysis by NHI Mgmt Group, based on content published by raxIT Labs: “The list finally caught the god agent”.

By the numbers:

• The 2026 OWASP LLM Top 10 was checked against 7,714 real incidents.
• The methodology classified 6,639 of those incidents.
• Excessive Agency moved from rank 6 to rank 3 in the 2026 list.

Key questions

Q: What breaks when an AI customer service agent has too much authority?

A: The control model breaks when the agent can both fetch sensitive context and act on it without strong policy boundaries.

Q: When should organisations prioritise runtime mediation over post-action review for AI agents?

A: As soon as an agent can touch production systems, external APIs, money movement, or sensitive data.

Q: What are the warning signs that an AI agent is overprivileged?

A: Warning signs include an agent that can reach tools or data stores unrelated to its task, persists with the same privileges after the session ends, or depends on static credentials instead of short-lived tokens.

Practitioner guidance

• Implement deterministic approval gates for high-impact actions Require every agent action that can move data, spend money, change permissions, or alter infrastructure to pass a policy decision point before execution.
• Reduce agent privilege to task scope only Issue the minimum tool access and data access needed for the current job, then revoke it when the task completes or the session ends.
• Separate reversible actions from irreversible actions Auto-approve only actions that can be cleanly undone, and route irreversible actions to human escalation or stricter policy checks.

Bottom line: The article frames excessive agency as the point where an AI system can turn delegated authority into an actual downstream action.

Explore further

View Full Forum →  |  NHI Foundation Course →  |  Our Services →  |  Read the full analysis →