Agentic AI
AI that plans, uses tools and acts across several steps: the projects, research and governance questions behind it.
What it is
Agentic AI describes systems that pursue a goal across several steps: they plan, call tools and APIs, check the result and adjust, instead of answering one prompt at a time. An agent might reconcile invoices, triage support tickets or research a market, and hand off to a person when it is unsure.
Why it matters to leaders
Agents move AI from advice to action, and that changes the risk profile. The questions that matter are about governance: which systems an agent can touch, what it may do without approval, how every action is logged, and how you measure whether it actually finished the task.
What to watch
Multi-agent orchestration, evaluation methods that test whole tasks rather than single answers, agent identity and permissions, and open protocols such as MCP that connect agents to tools in a standard way.
Latest agentic ai radars
Questions and answers
What is agentic AI?
AI systems that work towards a goal over several steps, deciding what to do next, using tools such as search, code or business applications, and checking their own progress, with people setting the goal and the limits.
How is an AI agent different from a chatbot?
A chatbot answers a message. An agent takes actions: it can open systems, run tools and complete a multi-step task, which is why it needs permissions, logging and human approval for consequential steps.
What are the main risks of agentic AI?
Acting on wrong information, permissions that are broader than the task needs, prompt injection through content the agent reads, runaway cost and unclear accountability. Least-privilege access, approval gates, full action logs and task-level evaluation address most of them.
Other beats in the village
- Quantum computingQubits, error correction, post-quantum security and the open-source SDKs where quantum developers are spending their time.
- AI solutionsThe applications built on top of models: retrieval, copilots, document AI and the open-source building blocks behind them.
- AI harnessesThe software around a model that turns it into a working agent: coding agents, terminals, sandboxes and the loop that runs them.
- Agent skillsPackaged know-how that teaches an agent to do a job well: SKILL.md folders, skill libraries and how teams share them.
- MCP serversConnectors that let any agent use a tool or data source through the Model Context Protocol, and how to adopt them safely.
- Agentic GitHubThe most-starred agent projects on GitHub and the releases they ship each week: a health check for the frameworks everyone builds on.