Skip to content 
New Open-Source Tool Accelerates Testing for Trustworthy AI in Luxembourg
The Luxembourg AI Factory has released the AI Assessment Sandbox Configurator, a new open-source tool co-developed by the Luxembourg Institute of Science and Technology (LIST) and the University of Luxembourg’s Interdisciplinary Centre for Security, Reliability and Trust (SnT), that helps organisations efficiently and rapidly configure and deploy AI assessment sandboxes tailored to their specific use cases.
Launched in June 2026, it allows any organisation to significantly reduce the time and cost needed to create, deploy and use environments for testing whether an AI system is trustworthy.
This release comes at a pivotal moment for AI adoption in Europe. Setting up a technical environment to assess AI can be complex and time-consuming: teams need to find relevant tests, set up sandbox environments, choose the right metrics, and bring results together into clear insights, all of which requires collaboration across technical, legal and business teams. The EU AI Act and other regulations add further obligations on AI providers to demonstrate trustworthiness, obligations that today remain slow, costly and largely manual to meet.
I. What the tool does
A shared Catalogue of AI Tests & Controls
The Catalogue is an open-source European asset co-developed by a broad community: a structured, extensible library of AI testing tools, controls and datasets, integrated through an open plug-in API interface. It is possibly to be endorsed by National Competent Authorities, with the potential to evolve into vertical marketplaces.
A configurable sandbox
By selecting tests, controls and datasets from the Catalogue that are relevant to their use case, organisations can generate a customised sandbox, complete with a dedicated collaborative dashboard and a report generator with templates suited to their specific needs.
Open source and community-built
The tool is available on GitHub, where organisations can install it in their own environment, extend the Catalogue with their own tests and controls via the plug-in interface, or contribute to the code of the Configurator itself.
II. Why it matters
Adapts to any AI use case
Every AI use case has different needs. The Configurator helps map an organisation’s specific context and requirements to the right tests, metrics and controls.
Brings results together and makes them comparable
All test results and control checks flow into one dashboard and report, with outputs aligned automatically for easy comparison, and all activity recorded in tamper-proof logs for auditability.
Aligns teams with a shared understanding
A common assessment structure helps multidisciplinary teams interpret results in the same way, making it easier to collaborate and hold clear discussions with stakeholders.
III. Looking ahead
Beyond its immediate use as a testing tool, the open and collaborative approach behind the AI Assessment Sandbox Configurator is intended to help build trust and support digital sovereignty through transparency and openness. For organisations navigating the EU AI Act, including private equity and venture capital firms integrating AI into their own operations or portfolio companies, the tool offers a practical, open-source way to move faster while keeping AI adoption trustworthy.
More information and the technical documentation are available on the dedicated Luxembourg AI Factory page: https://aifactory.lu/build-and-test/testing-the-solution/ai-sandbox-configurator, the Catalogue of AI Tests & Controls: https://sandboxconfigurator.aifactory.lu/catalogue, and the GitHub repository: https://github.com/lux-ai-factory/aisc
