Glossary of terms

An organisation that uses an AI system in its own activities. Most SMBs are deployers, and this is who Veillo is built for.

An organisation that develops an AI system or puts it on the market. Provider duties are heavier than deployer duties.

How the Act sorts a use: prohibited, high, limited, or minimal. The tier sets the obligations.

The Act's list of high-risk use categories, such as recruitment, creditworthiness, and biometric identification.

The technical documentation that should exist for a high-risk system, which a deployer keeps on file.

The Article 4 duty to make sure the people using AI on your behalf understand what it does and its limits.

General-purpose AI: broad models like large language models. They carry their own provider obligations under the Act.

Data protection impact assessment, a GDPR Article 35 step for processing of personal data likely to be high-risk.

Fundamental rights impact assessment, an Article 27 step required of certain deployers before using a high-risk system.

Classifications a lawyer has reviewed, which every customer using a tool that way inherits with the reasoning attached.

A model's suggested tier for a system Veillo has not seen before. It is always marked and confirmed by a person.

A weighted measure on your dashboard of how many of your duties are in place, with high-stakes duties counting for more.