Agentic AI
Also known as: AI agents, autonomous agents, agent systems
Agentic AI refers to AI systems that autonomously plan, decide, and execute multi-step actions to achieve a goal, typically by combining a language model with tools, memory, and feedback loops.
Detailed explanation
Agentic AI is a class of AI systems where a model is given a goal and the authority to take actions toward it — rather than just returning a single response. The system typically iterates: it decides what to do next, invokes tools (APIs, databases, code execution), observes results, and updates its plan.
Core building blocks include a foundation model (the “reasoner”), a tool layer (functions the agent can call), memory (short-term context plus longer-term storage), and a control loop that bounds time, cost, and risk. Multi-agent variants coordinate several specialized agents through a planner or orchestrator.
In enterprises, agentic AI is being applied to customer support deflection, internal operations (HR, finance, IT triage), data analysis, and software engineering tasks. The hard parts in production are evals, guardrails, cost control, and graceful human handoff — not the agent loop itself.