Excessive Agency Examples?
7.0BRAINROT SCOREExcessive Agency Granting an AI agent more autonomous permission, tool access, or decision-making latitude than its actual task requires, so a single successful manipulation can cause damage far beyond what the task itself should ever have allowed.
Origin:One of the OWASP-catalogued LLM application risk categories, describing a structural over-permissioning problem rather than any single attack technique.
First Seen:2023
Peak Era:2026 (Agent Security)
Aura Impact:+1 Aura (Scoping an Agent's Permissions to Exactly What the Task Needs) / -2 Aura (Giving an Agent Admin Access Because Scoping It Down Was Annoying)
EXAMPLE USAGE
"Nobody meant to grant the agent delete permissions on the whole bucket — pure excessive agency, discovered the hard way."
Want the full breakdown — categories, trend velocity, platform distribution, and community voting on Excessive Agency? Visit the full dictionary entry for Excessive Agency.