AI search is making biggest impact on businesses since the Internet first became widely adopted.
The important change is not simply that AI helps people find products. Increasingly, AI systems are making product and technology choices as part of a larger task. They may search for options, compare them, select one, and incorporate that choice into a plan before a human reviews the result.
By the time the human becomes involved, many of the underlying choices may already have been made. Asking the AI to reconsider those choices or evaluate a different set of products requires additional effort.
I’ve been trying to better understand this process by experimenting with some of the same engineering techniques used to build retrieval-augmented generation (RAG) systems. They provide a useful conceptual framework for understanding why relatively small changes to a piece of content can affect whether an AI system retrieves it, considers it relevant, and ultimately surfaces it in response to a query.
Oppvera is a free educational service @jcasman and I built to help people explore these ideas and better understand how AI search works.
Here’s a short video showing the approach:
If you try it, I’d be interested to know whether it improves your understanding of AI search.