Agentic AI
Agentic AI refers to AI systems that can pursue a goal with limited human direction by planning steps, using tools or data, taking actions, and adjusting based on the results.
Most generative AI tools respond to a single prompt and stop. An agentic system works toward an outcome instead. Given a goal, it breaks the work into steps, decides what to do next, calls tools or other systems, checks what happened, and keeps going until the task is done or it needs a person.
The term covers a range of autonomy. Some agents only assist one user with a narrow task. Others run inside a defined business process, and a smaller number act on their own across several systems. Because agents can take real actions, like updating records or sending messages, questions about access, ownership, and oversight matter as much as the model behind them. Agentic AI is also different from traditional automation, which follows fixed rules. An agentic system can choose its own path, so it needs clear limits and a way to escalate.
Key parts
Goal. A defined outcome the system works toward.
Planning. Breaking the goal into steps and choosing what to do next.
Tool use. Calling applications, data sources, or other agents to act.
Memory and context. Keeping track of what has happened so far.
Feedback. Checking results and adjusting the plan or asking a person for help.
