The table below defines many of the terms related to AI agents and provides links to more information about them.
Related article:
| Term | Definition | Learn more |
| Advanced AI agent | A previous term for an AI agent created in the AI agents - Advanced add-on. On May 11, 2026, the AI agent functionality previously restricted to the add-on became part of all Suite and Support plans. | About AI agents |
| Agentic AI | A type of artificial intelligence that can perceive, reason, plan, and act independently to achieve complex goals with minimal human supervision. | About AI agents with agentic AI |
| AI agent | An AI-powered chatbot that interacts with customers on one of your support channels. | About AI agents |
| AI agent ticket | A ticket for a messaging conversation that was handled entirely by an AI agent or a default messaging response. By definition, an AI agent ticket hasn’t yet been or never was escalated to a human agent. | Understanding AI agent tickets for AI agent–only conversations |
| Automated resolution | The unit of measurement used for calculating and billing your account for AI agent usage. | About automated resolutions for AI agents |
| Configured action | A configurable task that an AI agent can perform during a conversation with a customer. | About configured actions for advanced AI agents |
| Conversation flow | How a conversation between an AI agent and a customer unfolds. A conversation flow can be shaped by a generative procedure or a dialogue. |
Creating generative procedures to shape flexible conversation flows for AI agents Creating dialogues to define scripted conversation flows for AI agents |
| Default messaging response | A previous method of configuring the first message sent to a customer when they launch the Web Widget. This response included a basic greeting, a request for more information, and a message letting the customer know they’re being connected to a human agent. | Configuring messaging responses and business hours |
| Dialogue | A scripted conversation flow that uses branching logic to determine an AI agent's responses and actions during a conversation with a customer. | Creating dialogues to define scripted conversation flows for AI agents |
| Entity | A piece of information that has a specific meaning, which an AI agent can recognize and treat in a particular way. | Creating entities in conversation flows for advanced AI agents |
| Escalation | The process of handing off a customer conversation from an AI agent to a human agent. | Configuring escalation strategies and flows for advanced AI agents |
| Essential AI agent | A previous term for an AI agent created using the AI agents - Essential level of AI agent functionality. On May 11, 2026, AI agents - Essential became legacy functionality. | About AI agents |
| Expression-based AI agent | A previous generation of Zendesk AI agents that used manually trained expressions to understand customer requests and link to scripted dialogues. | About AI agents |
| Generative AI | A type of artificial intelligence capable of creating content based on large training datasets. In AI agents, the created content types include text and audio. | Generative AI at Zendesk |
| Generative procedure | A flexible conversation flow that uses agentic AI to guide an AI agent's responses and actions during a conversation with a customer. | About generative procedures for AI agents |
| Generative reply | Content that an AI agent creates using generative AI. Also referred to as an AI-generated answer. | Configuring generative replies in AI agents |
| Integration builder | A no-code tool that enables you to connect an AI agent to any API or data source without extended technical or programming skills. | About the integration builder for AI agents |
| Legacy AI agent | A term for an AI agent created using AI agent functionality that is no longer current. Often refers to an AI agent built using the bot builder, answers, and intents. | About AI agents |
| Parameter | A piece of information you can collect and use to tailor AI agent workflows, including using in conditional blocks, segments, and API integrations. | Using parameters in AI agents |
| Procedure | See Generative procedure. | About generative procedures for AI agents |
| Reply | A container for the dialogue that's triggered when the associated use case is detected during a conversation. Distinct from Reply method. | Creating and managing replies for AI agents |
| Reply method | Whether a use case uses a generative procedure or a dialogue. Distinct from Reply. | Configuring whether a use case uses a dialogue or generative procedure |
| Segment | A collection of customers who share common traits. You can use segments to tailor AI agent workflows for distinct subsets of your user base. | Creating segments to target specific customers in AI agent conversations |
| Structured message template | A form, webview, or custom template that can be used to collect or present certain information to a customer during a conversation with an AI agent. Distinct from Template. | Creating and managing structured message templates for AI agents |
| System reply | A response that an AI agent sends to a customer at a specific point in a conversation, such as a welcome message or a feedback request. | Viewing and editing system replies for AI agents |
| Template | A centralized reply that you can link to from multiple generative procedures or dialogues. Distinct from Structured message template. | Using templates in conversation flows for AI agents |
| Use case | The mechanism by which an AI agent understands what a customer is asking about during a conversation. Use cases are the topics that your customers need help with, such as order returns or refund requests. | Creating use cases for AI agents to identify what customers are asking about |
| Zero-training AI agent | A previous generation of Zendesk AI agents that used use cases to understand customer requests and link to scripted dialogues. Much of the configuration performed for zero-training AI agents (such as use cases) remains the same for the current, agentic AI generation of Zendesk AI agents. The important distinction is that AI agents with agentic AI can also use generative procedures. | About AI agents |