Export
The link always opens the latest version of this page.
Ask an AI about it
Open in ChatGPTA new chat that reads this entry from the site Open in ClaudeThe same, in ClaudeShare the link instead of a file, and they will always see the latest version.
Files are a snapshot of today. The live page has anything newer.
Interactive AI Agent Pattern
User uses interactive AI client to determine actions with user available for consent
The Interactive AI Agent Pattern is a technique in which a user employs an interactive AI client to determine and authorize actions, with the user remaining available to provide consent throughout the execution process.
Overview
This pattern represents a human-in-the-loop approach to AI-assisted operations, where decision-making authority remains with the user rather than being fully delegated to automated systems. The user maintains active availability to review, approve, or reject proposed actions before they are executed.
Detection Characteristics
The presence of this pattern can be identified by two key indicators:
- User availability for consent: The user is actively present and available to provide consent when required
- Conditional consent availability: The user remains available to provide consent if needed, even if not continuously monitoring
These characteristics distinguish interactive AI agent patterns from fully autonomous systems where users have limited or no opportunity to intervene.
Related Concepts
See also: AI Agent, Human-in-the-Loop Security, Consent-Based Access Control
Attack ID: interactive-agent