AI customer support glossary
Plain-language definitions of the concepts behind modern AI customer support: what they mean, how they fit together, and why they matter for support teams. Each entry is written to be accurate and vendor-neutral, with practical context for anyone building or evaluating an AI support agent.
Ticket deflection
Ticket deflection is the practice of resolving a customer's question before it becomes a support ticket (usually through self-service like help articles, FAQs, or an automated agent), so an agent never has to handle it manually.
RAG chatbot
A RAG chatbot is a conversational AI that uses retrieval-augmented generation: it looks up relevant information from a knowledge source at answer time and uses that retrieved content to ground its response, rather than relying only on what the model memorized during training.
Customer service metrics
Customer service metrics are the quantitative measures teams use to track the quality, speed, and efficiency of their support (things like resolution rate, first response time, CSAT, and ticket volume) so they can spot problems and improve the customer experience.
AI grounding
AI grounding is the practice of tying an AI model's responses to verifiable, authoritative information, such as your own documents or live data, so its answers reflect real facts rather than guesses, reducing hallucinations.
AI agent
An AI agent is a software system that uses an AI model to understand a goal, decide what to do, and take actions toward it, often across multiple steps and tools, rather than just returning a single canned response.
Conversational AI
Conversational AI is technology that lets software understand and respond to human language in a natural, back-and-forth dialogue, powering chatbots, voice assistants, and AI support agents across text and voice.
Put these ideas to work
Build an AI support agent that resolves on your knowledge, live in an afternoon.
No credit card required