Category drift is a digital readability reference point related to how AI systems reconstruct an organisation. The term helps move from vague concern to qualifiable observation.
Why the term matters
An organisation does not only lose visibility when it is misread by AI systems. It can lose precision, category, recommendability or coherence. This term helps name one of those movements so the right correction can be selected.
In Pagup’s vocabulary, it is not an isolated keyword. The term must be connected to a baseline, canonical sources, proof and a correction trajectory.
What the term does not mean
It does not describe a simple variation in wording. Two answers may use different words while keeping the same understanding. The problem begins when the substance changes: wrong category, old offer, wrong competitor, lost differentiation or weakened recommendation.
How to use it
In a diagnostic, this term classifies an observed gap. It helps determine whether the work should target content, brand, proof, internal links, AI governance or a deeper rebuild.
For a deeper doctrinal reference, see the corpus on gautierdorval.com.