Data and benchmarks: what our conclusions rest on

Our conclusions rest on the test bench, the incident chronicle and benchmark sets, each published with its composition and its limits. The data is open; a customer’s telemetry never is.

Data without a register is a warehouse without a list Data without a register is a warehouse without a list
Data without a register is a warehouse without a list

Other people’s figures are cited with links; our own measurements are reported separately and with the conditions of the run. The bench, the chronicle and the benchmark sets are working instruments rather than a shop window.

Data is published under CC BY 4.0 unless stated otherwise: chronicle exports as CSV, methodologies as text. Closed data — a customer’s telemetry — is never published, and that is a condition of the work rather than a preference.

Planned releases are listed as planned on the datasets page. We do not present a plan as a result, and we mark what does not exist yet, and why it is not there.

Materials on this topic

Other topics

AI red teaming: what a test shows, and what it does notAI standards: where the rules actually applyAI supply chain: what enters before the model runsMulti-agent systems: behaviour between the agentsIncident chronicle: documented consequences, not forecastsBehavioural audit: how agent behaviour drifts over timePrompt injection: where text becomes an instructionRussian practice: who is held responsible for AIMethodology: how the Institute tests AI systemsSociology of AI: relations between the agentsGovernment customers: how to commission a testGovernance and funding of the InstituteCases: scenarios of work and what they show

Page updated 02 November 2026 · site build 2026.11.13 · ANO «ISAIS»