For decades, protecting privacy has meant one thing: stop personal data from leaking out. Gartner thinks that idea is about to break.
By 2029, most privacy incidents will come not from leaked personal data but from what AI infers about people, the research firm predicted this week. The risk is no longer only the records a company holds. It is the conclusions a model can draw from scraps that look harmless.
From data exposure to insight exposure
Gartner analyst Bart Willemsen calls it a shift “from data exposure to insight exposure.” A model can now reconstruct deeply personal things, like a health condition or a behavioural pattern. It pulls them from data that looks anonymous, aggregated or harmless, and never has to breach a database to do it.
There is an irony in this. Companies are storing less personal data, pushed by regulation and cost. That is meant to reduce risk. But if an AI can infer the sensitive detail anyway, holding less data does little to protect the person it describes.
