One number, 0–100. Lower is safer. It turns a vague fear — “what does AI say about me?” — into something you can measure, track, and improve.
Your score is a weighted blend of five things, each scored from what the models actually said about you.
How much of your real personal data the models surface: home address, phone, employer, family. This is weighted heaviest because it’s the scariest and the most actionable.
False, negative claims — weighted by how confidently the models state them and how many models repeat them. A lie that appears everywhere hurts more than a one-off.
How badly you’re blended with strangers who share your name, and wrong biographical facts presented as yours.
When models cite a source about you, is it one you own (your site, your profiles) or a data broker, a smear, or nothing at all?
How much the answer swings between models and between runs. An unstable answer is its own kind of risk.
Your answer is accurate, sparse, and low-risk. The job now is monitoring so it stays that way.
Models are blending you with others or repeating at least one false or exposing claim. Fixable — and worth fixing before it hardens.
Multiple models expose personal data or confidently repeat false claims. This needs priority attention.
You can’t manage what you can’t measure. The score gives you a single, honest read today — and a line that visibly drops as we do the work. It’s how you know the problem is real, and how you know it’s getting handled.