The technology behind
every Impersonato test.
Every Impersonato test runs on the same controlled technology. We call it Impersonato Cognitive Runtime.
A model gives an answer.
A service needs control.
A general-purpose model can generate an answer. A test you can rely on needs more: the dummy has to be the right person, do a defined task, use only the data it should, keep a record of what happened, and know when to hand a case to a person.
That is what Impersonato Cognitive Runtime provides — for every test we build.
Four things we set for
every service.
The person the dummy stands in for
The customer or user whose point of view the dummy takes — built as a detailed profile, not a one-line instruction.
What it does, step by step
The steps of the work and the shape of the result, so the same brief produces the same kind of output every time.
What it may access, and when it must stop
Data-access rules, usage limits, optional approval gates, and a clear point where an uncertain case goes to a person.
Where it runs, and who controls it
Run through us, in your own environment with your model accounts, or in a dedicated private instance — with a record of what happened.
One runtime,
two kinds of work.
Marketing & Communication
The Runtime runs interviews, panels and message tests with dummies of your audience, compares their reactions, and produces a clear report — with every run logged so the findings can be traced back to what was actually asked.
AI Safety
The Runtime runs vulnerable dummies against your AI in long, back-and-forth conversations, scores each one against a clear rubric, and returns a list of what failed and how to fix it — within the limits you set.
Built as cognitive
architecture.
A role in the Runtime is not a styled prompt. It is a constructed cognitive architecture — an identity, a psychological structure, a characteristic way of attending to a problem, and a way of deciding under uncertainty.
Assembled this way, the model reasons from a coherent point of view instead of collapsing into balanced, hedged, agree-with-everything output. A simulated respondent holds genuine positions, brings its own priorities and suspicions, and shifts under pressure the way a person does — rather than flattering whatever it is shown.
Inner life, not demographics
Identity, motivation and stance — the layer that gives a simulated person actual opinions, where a demographic label on its own leaves a model hedging.
State that changes over a conversation
A persona reacts to warmth, provocation and pressure across turns — the same exchange can move it, the way real interactions move people.
The system doesn't grade itself
Construction, simulation and judgement run on separate models, so a role isn't scoring its own work or leaking its own preferences into the result.
Measured, not asserted
Behaviour is calibrated against real-world data and read as distributions across many runs — never a single, cherry-picked output.
We hold this to an evidence standard. In controlled ablations, keeping the surface approach but removing the underlying cognitive mechanisms collapses the output straight back to generic. The depth does measurable work — and preserving it in production, run after run, is what the Runtime is built to do.
Run it the way you
require.
Managed by Impersonato
We run the workflow and deliver the result.
Customer-controlled
The workflow runs with your model accounts and your usage limits.
Private deployment
For sensitive environments, the Runtime can be deployed in a dedicated or isolated setup.
See how our systems
are controlled.
Whether you test your marketing or stress-test your AI, both run on the same controlled technology. Tell us which one you need.