Generative AI generates. Agentic AI acts.
Pessoal, os dois usam o mesmo modelo. A diferença não está no README. Está no que o turno fez . Você já teve aquele chat em que o “agente” resolveu tudo em Markdown e o git tava igual? Pois é. Eu tava na aula Introdução à IA agêntica , módulo 1 do Agentic AI with LangChain and LangGraph , no Coursera (IBM). A irmã LangGraph versus LangChain virou o denso de cadeia, grafo e harness . Aqui é mais…
The article discusses the difference between generative AI and agentic AI, using LangChain and LangGraph. Generative AI takes a prompt and produces text, whereas agentic AI chooses the next step, interacts with tools, and has consequences. The author highlights that generating text is not the same as taking action, and that a true agent involves a loop of observing, deciding, and using tools. The article also touches on the importance of human involvement, security, and auditing in agentic AI systems.
Written by urgent.news from Dev.to's report — not a translation of it. Machine-written — may contain errors; check the original before relying on it.