Dashboard speeds up responsible AI checks for large language models
VERA: A Responsible AI Dashboard for Assessing LLM Deployments in Light of the EU AI Act
Software EngineeringArtificial IntelligenceComputers and Society
Summary
Organizations using large language AI need to prove their systems are safe and fair, especially under EU rules. The authors created VERA, a dashboard that helps non-technical reviewers run tests on these AI models and see clear scores on different safety and fairness aspects. This tool also tracks the energy used during testing, showing the environmental impact. People in a bank used VERA and found it faster and easier to understand than reading raw test results, cutting the checking time almost in half.
What this means in practice
- •For compliance teams: Run and visualize AI model tests to support regulatory approval under the EU AI Act.
- •For ai operations teams: Quickly assess multiple responsible AI requirements and resource usage during model deployment decisions.
Authors
Robin Quériaux, Solveig Bachellery, Léa Déléris, Saurabhsingh Rajput, Mariam Barry, Tushar Sharma, Jérôme Lebecq
Abstract
Institutions that deploy large language models in regulated settings often need technical evidence of responsible AI practices before approving a deployment. COMPL-AI provides a technical interpretation of the EU AI Act that organises this evidence under six principle groups and eighteen technical requirements, twelve of which current benchmarks can measure. The benchmark suites that produce this evidence are typically operated from a command line, while the staff who approve deployments are non-engineers. We present VERA, an open-source self-hosted dashboard deployed to facilitate responsible AI processes within a large European bank. A reviewer selects a model and runs public benchmarks against it. VERA scores the twelve requirements that current benchmarks can measure and instruments the energy consumed by each evaluation, adding run-level evidence for environmental impact. We run four served open-weight models through the requirement pipeline, and the dashboard displays a score for each measurable requirement. A within-subject study then compares how participants read these benchmark results: from the raw output files in one condition, and through the VERA dashboard in the other. The dashboard reduced the time needed to perform the responsible AI assessment, reducing the time required from 56 seconds using raw files to 30 seconds using our dashboard, on average. Our dashboard scored highly (4.36 out of 5) on usefulness in the survey conducted within the organization. The dashboard and the study material are released publicly to promote replicability and reproducibility in different organisational contexts.