Grounding
Giving an AI real source material to answer from, instead of letting it rely on memory. A grounded answer can be checked and cited; an ungrounded one is where hallucinations come from.
Related terms
Predictive Analytics
Using historical data to estimate what will happen next: which customers will lapse, which stock will run out. Useful precisely to the degree that the past resembles the future.
AI Overviews
The written summary Google now places above its search results, assembled from several pages and citing a few of them. Appearing inside one can bring visitors; being left out of one can quietly cost you the clicks you used to get.
Microdata
An older way of marking up facts, written as attributes on the HTML itself rather than in a separate block. Still understood, but JSON-LD is now the safer choice.
Unsupervised Learning
Training a model on unlabelled data and letting it find structure by itself — grouping similar customers, say, without being told what the groups are.
User-Agent
The name a browser or bot gives when it requests a page. It is how a server tells Googlebot from ChatGPT from a person, and how robots.txt rules are addressed to one and not another.
Model Drift
When a model gets worse over time because the world moved on and it did not. A pricing or demand model trained two years ago quietly stops matching how customers behave now.