Trust boundaryThe one you cannot see

Your team trusts every AI tool in this lineup.
One is moving data out.

Shadow AI, personal accounts, and agents that reach into your systems all look like sanctioned work. Run the sweep on your own stack in fifteen questions and see the exposure path before someone else does.

02 / THE PROBLEM IN THREE NUMBERS
Bring-your-own AI
0%

of the people using AI at work bring their own AI tools.

Source: Microsoft & LinkedIn, Work Trend Index 2024
Personal accounts
0%

of employees using AI on corporate devices sign in with a personal account.

Source: Verizon DBIR 2026
Unseen use
0%

of organizations have evidence or suspect employees use public generative AI at work.

Source: Gartner 2025
What the Index measures

Five weighted signals, one exposure score.

The Index scores where AI-era data risk actually lives, from whether you can see AI use at all to whether you can act on it. Each signal carries a weight; together they resolve to a single reading from 0 to 100.

04 / THE EXPOSURE SCALE
The reading

From Contained to Critical.

Your answers resolve to one of five bands. The band is a starting point, not a grade: it tells you how far AI-era data can move before something stops it.

020406080100
45
ContainedGuardedElevatedExposedCritical
Elevated

Real exposure sits in more than one signal. This is where most teams are today.

Drag to explore · take the assessment for your real score
15 questions · ~4 min · no sign-up

Get your reading in about four minutes.

Answer fifteen questions across the five signals and see your exposure score and band immediately. No registration to see your result.

Run the assessment

Runs entirely in your browser. Nothing is sent anywhere until you ask for the report.

0 ELEVATED
Sample reading
06 / METHOD

A vendor-neutral instrument.

The Index weights the two signals where AI-era data actually leaks, visibility and movement, most heavily, because you cannot govern what you cannot see. Every figure on this site comes from a major research source and is cited where it appears. No vendor is named in the scoring.

GARTNERVERIZON DBIRMICROSOFTIBMPONEMON
Full methodology
Signals5
AI visibilityweight 25
Data movementweight 25
Identityweight 20
Evidenceweight 15
Remediationweight 15
Bands5 · 0–100
Questions15
A vendor-neutral read on AI-era insider risk

See how exposed you are.

Fifteen questions, about four minutes, your result in your browser.

Run the assessment