Reference

AI Chatbot Glossary

From RAG to NLP, learn the key terms and definitions behind AI chatbots, conversational AI, and customer automation technology.

AI Agent

An autonomous software program powered by artificial intelligence that can perceive its environment, make decisions, and take actions to achieve specific goals. AI agents go beyond simple chatbots by handling multi-step tasks, using tools, and adapting their behavior based on context.

See AI Agents in action

AI Chatbot

A conversational interface powered by artificial intelligence that can understand natural language, generate human-like responses, and assist users with tasks such as answering questions, capturing leads, or processing orders. Unlike rule-based bots, AI chatbots learn from context and improve over time.

Build your AI chatbot

API (Application Programming Interface)

A set of protocols and tools that allows different software applications to communicate with each other. In the chatbot world, APIs enable your chatbot to connect with CRMs, e-commerce platforms, helpdesks, and other business tools to exchange data and trigger actions.

Explore Chatonbo API docs

Auto-Reply

An automated response sent by a chatbot or messaging system without human intervention. Auto-replies can be triggered by specific keywords, user actions, or time-based rules to provide instant acknowledgment or answers to common queries.

Bot

Short for "robot," a bot is a software application that runs automated tasks. In customer communication, bots handle repetitive interactions like answering FAQs, routing inquiries, and collecting contact information, freeing up human agents for complex issues.

Bot Training

The process of providing a chatbot with data, instructions, and examples so it can respond accurately to user queries. Training can involve uploading knowledge base documents, writing custom instructions, setting conversation tone, and fine-tuning responses based on real interactions.

Learn how to train your bot

Bounce Rate

The percentage of website visitors who leave a page without interacting or navigating further. Chatbots can significantly reduce bounce rates by proactively engaging visitors with targeted messages, offers, or assistance before they leave.

Chatbot

A software application designed to simulate human conversation through text or voice. Chatbots range from simple rule-based systems that follow predefined scripts to advanced AI-powered agents that understand context, learn from interactions, and handle complex conversations.

Create a free chatbot

Chatbot Widget

An embeddable UI component that adds a chat interface to any website. Typically appearing as a floating button in the corner of a page, the widget opens into a chat window where visitors can interact with your AI chatbot without leaving your site.

See the Chatonbo widget

Conversational AI

A branch of artificial intelligence focused on enabling machines to understand, process, and respond to human language in a natural, contextual way. Conversational AI combines NLP, machine learning, and dialog management to power chatbots, virtual assistants, and voice interfaces.

Conversion Rate

The percentage of visitors who complete a desired action, such as making a purchase, filling out a form, or subscribing to a service. AI chatbots improve conversion rates by engaging visitors at the right moment, answering objections, and guiding them toward a decision.

Boost conversions with AI

CRM (Customer Relationship Management)

A system for managing all your company's interactions with current and potential customers. Chatbots integrate with CRMs like HubSpot, Salesforce, and Zoho to automatically sync captured leads, log conversations, and update contact records in real time.

View CRM integrations

Customer Support Automation

The use of technology, including AI chatbots, to handle customer service inquiries without human intervention. Automation can resolve common issues instantly, route complex tickets to the right team, and provide 24/7 support across multiple channels.

Automate your support

Decision Tree

A flowchart-like structure used to guide chatbot conversations through a series of predefined questions and branches. Each user response leads to a different path, making decision trees useful for troubleshooting, qualification, and guided workflows.

Deep Learning

A subset of machine learning that uses artificial neural networks with multiple layers to learn complex patterns from large amounts of data. Deep learning powers the language understanding capabilities of modern AI chatbots, enabling them to comprehend nuance, context, and intent.

Embedding

A numerical representation of text (words, sentences, or documents) in a high-dimensional vector space. Embeddings allow AI systems to measure semantic similarity between pieces of text, which is essential for knowledge base search, intent matching, and retrieval-augmented generation.

Exit Intent

A technology that detects when a website visitor is about to leave the page (e.g., moving the cursor toward the browser's close button). Chatbots can use exit intent triggers to display a targeted message, offer, or conversation prompt to re-engage the visitor before they bounce.

FAQ Bot

A chatbot specifically designed to answer frequently asked questions. FAQ bots pull answers from a pre-built knowledge base and can handle a high volume of repetitive inquiries, reducing the load on human support teams while providing instant responses to customers.

Build an FAQ bot

Flow Builder

A visual tool for designing chatbot conversation flows using a drag-and-drop interface. Flow builders let non-technical users create complex conversation paths, set up conditional logic, and define actions like sending emails or updating CRM records.

Generative AI

A category of artificial intelligence that can create new content, including text, images, code, and more. In chatbots, generative AI (powered by models like GPT-5.2, Claude, and Gemini) enables dynamic, context-aware responses rather than relying on pre-written scripts.

GPT (Generative Pre-trained Transformer)

A family of large language models developed by OpenAI that can understand and generate human-like text. GPT models are widely used to power AI chatbots, enabling them to hold natural conversations, answer questions, summarize information, and assist with complex tasks.

Hallucination

When an AI model generates information that sounds plausible but is factually incorrect or fabricated. Hallucinations are a known challenge in generative AI chatbots and can be mitigated through techniques like RAG, grounding responses in verified knowledge bases, and setting strict instructions.

Human Handoff

The process of transferring a chatbot conversation to a live human agent when the bot cannot resolve an issue or the user requests human assistance. Effective handoff preserves conversation context so the agent can continue seamlessly without asking the customer to repeat themselves.

Intent Recognition

The AI capability of identifying what a user wants to accomplish from their message. For example, recognizing that "I want to cancel my subscription" indicates a cancellation intent. Accurate intent recognition is critical for routing conversations and providing relevant responses.

Integration

A connection between your chatbot and an external tool or platform that enables data sharing and automated workflows. Common integrations include CRMs (HubSpot, Zoho), helpdesks (Zendesk), messaging apps (Slack, WhatsApp), and e-commerce platforms (Shopify).

Browse integrations

Knowledge Base

A structured collection of information, such as documents, FAQs, product details, and policies, that a chatbot uses to find accurate answers to user questions. AI chatbots search the knowledge base in real time to provide grounded, relevant responses instead of guessing.

Learn about knowledge bases

Large Language Model (LLM)

An AI model trained on vast amounts of text data that can understand and generate human language. LLMs like GPT-5.2, Claude, and Gemini power modern chatbots, enabling them to hold natural conversations, follow complex instructions, and reason about information.

Lead Capture

The process of collecting contact information (name, email, phone number) from website visitors during a chatbot conversation. AI chatbots can naturally weave lead capture into the conversation flow, asking for details at the right moment without feeling intrusive.

Capture more leads

Lead Qualification

The process of evaluating whether a potential customer is a good fit for your product or service. AI chatbots qualify leads by asking targeted questions about budget, needs, timeline, and authority, then scoring or routing them based on the responses.

Qualify leads automatically

Live Chat

A real-time communication channel that allows website visitors to chat with human agents directly. Many businesses combine live chat with AI chatbots, using the bot to handle initial inquiries and routing complex cases to live agents via human handoff.

Machine Learning

A branch of AI where systems learn and improve from experience without being explicitly programmed. Machine learning algorithms analyze patterns in data to make predictions, classify information, and improve chatbot responses over time based on real interactions.

Multi-Language Support

The ability of a chatbot to understand and respond in multiple languages. This feature allows businesses to serve a global audience without building separate bots for each language, with the AI automatically detecting and responding in the visitor's preferred language.

Go multilingual

Multilingual Chatbot

A chatbot that can converse fluently in two or more languages, automatically detecting the visitor's language and responding accordingly. Multilingual chatbots eliminate language barriers and enable businesses to provide consistent support across international markets.

See multilingual features

Natural Language Processing (NLP)

A field of AI focused on enabling computers to understand, interpret, and generate human language. NLP powers chatbot capabilities like intent recognition, entity extraction, sentiment analysis, and language translation, making conversations feel natural and intuitive.

Natural Language Understanding (NLU)

A subset of NLP that focuses specifically on machine reading comprehension, enabling chatbots to understand the meaning behind user messages rather than just matching keywords. NLU handles nuances like synonyms, context, typos, and implied meaning.

Omnichannel

A customer engagement strategy that provides a seamless experience across multiple communication channels, including website chat, WhatsApp, email, social media, and SMS. Omnichannel chatbots maintain conversation context as customers switch between channels.

Open-Ended Conversation

A chatbot interaction style where users can type anything in their own words, rather than selecting from predefined options. AI-powered chatbots excel at open-ended conversations because they can understand and respond to a wide range of inputs using natural language processing.

Personalization

The practice of tailoring chatbot responses and experiences based on individual user data, such as browsing history, past purchases, location, or previous conversations. Personalized chatbots create more relevant interactions that increase engagement and conversion rates.

Prompt Engineering

The art and science of crafting instructions (prompts) that guide AI models to produce desired outputs. In chatbot development, prompt engineering involves writing system instructions that define the bot's personality, knowledge boundaries, response format, and behavioral rules.

RAG (Retrieval-Augmented Generation)

A technique that enhances AI responses by first retrieving relevant information from a knowledge base, then using that information as context when generating a response. RAG reduces hallucinations and ensures chatbot answers are grounded in accurate, up-to-date data from your documents.

See how RAG works at Chatonbo

Response Time

The time it takes for a chatbot to reply to a user message. AI chatbots typically respond in 1 to 5 seconds, dramatically faster than human agents. Low response times improve customer satisfaction and keep visitors engaged in the conversation.

Sentiment Analysis

The AI capability of detecting the emotional tone behind a user's message, such as whether they are happy, frustrated, confused, or angry. Chatbots use sentiment analysis to adjust their responses, escalate to human agents when frustration is detected, or prioritize urgent issues.

Session

A single continuous interaction between a user and a chatbot, from the first message to the end of the conversation. Sessions track conversation history, user context, and captured data. A new session typically begins when a visitor returns after a period of inactivity.

Smart Reply

AI-generated response suggestions that appear as clickable options during a chatbot conversation. Smart replies anticipate what the user might want to say next, making interactions faster and reducing typing effort, especially on mobile devices.

Token

The basic unit of text that AI language models process. A token can be a word, part of a word, or a punctuation mark. For example, "chatbot" is one token, while "AI-powered" might be three. Token counts affect API costs and the maximum conversation length a model can handle.

Training Data

The information used to teach an AI model how to respond. For chatbots, training data includes knowledge base documents, website content, FAQs, product catalogs, and conversation logs. The quality and relevance of training data directly impacts the accuracy of chatbot responses.

Trigger

A condition or event that causes a chatbot to initiate a specific action, such as sending a welcome message, offering help, or displaying a promotion. Common triggers include page load, time on page, scroll depth, exit intent, and specific URL visits.

Vector Database

A specialized database designed to store and search high-dimensional vector embeddings efficiently. Vector databases power the semantic search capability of RAG-based chatbots, enabling them to find the most relevant knowledge base content for any given user query.

Voice Bot

A chatbot that interacts with users through spoken language rather than text. Voice bots use speech recognition to understand spoken input and text-to-speech to deliver responses, enabling hands-free customer interactions via phone systems, smart speakers, and voice assistants.

Webhook

An HTTP callback that sends real-time data from one application to another when a specific event occurs. Chatbots use webhooks to notify external systems about events like new leads captured, conversations completed, or specific keywords mentioned, enabling automated workflows.

Set up webhooks

Widget

A small, self-contained application component that can be embedded into a website. A chatbot widget is the visible chat interface that visitors interact with, typically added to a site with a single line of code (a script tag) and customizable to match the site's branding.