Ask "generative or agentic?" and the honest answer is often "both, in that order." Generative AI produces content on request - text, images, code, summaries. Agentic AI uses that ability to do things: it strings together several steps, uses tools and works towards a goal with limited supervision. One is the engine; the other is the driver.
The simplest way to tell them apart
- Generative AI answers a question or produces a draft, then stops and waits for you
- Agentic AI takes a goal and keeps going - deciding the next step, acting on it, checking the result
If your problem is "help me write this," that is generative. If it is "handle this whole task for me," that is agentic. For a fuller contrast with older tools, see agentic AI vs traditional automation.
Which to start with
Most businesses should start with generative wins because they are cheaper, faster and lower-risk - drafting, summarising, answering FAQs, tidying data. Our guide to what AI can do for your business right now is a good map of those. Move to agentic once you have a repeatable, well-understood process worth automating end to end.
A quick test
Ask three questions about the task:
- Does it repeat often enough to justify setup effort?
- Are the steps clear enough to write down as rules?
- Can a mistake be caught before it causes real harm?
Three yeses point to an agentic use case. Fewer than that, and a generative tool with a human in the loop is the safer bet.
The bottom line
Generative AI writes; agentic AI acts. You will very likely use both - the question is not which to buy, but which fits each task. Learning to make that call deliberately is exactly what the Strategic Application of AI in Business course at London School of Business UK develops. Enquire today.