Agentic AI
What is agentic AI?
Agentic AI describes AI systems that can pursue a goal over several steps: planning what to do, using tools and business systems, checking results and deciding the next action, rather than answering a single prompt.
In customer service, an agentic AI might take a request such as “change my delivery address”, look up the order in the order system, check the policy, make the change through an API, confirm it with the customer and record the interaction in the CRM, handing over to a human agent if something falls outside its limits.
Because agentic AI takes actions rather than only producing text, the controls matter as much as the model: which systems it may touch, what it may change without approval, how it hands off to a person with full context, and how every step is logged. Contact centers usually start with narrow, well-defined tasks and widen the scope as results are proven.
Related: AI agents, Human in the loop, AI guardrails.