AI Readiness and Assessment

People

Publications (2024)

Projects

The AIRA Research Group operates within the Human-centric AI, Data and Software (HANDS) unit at the Luxembourg Institute of Science and Technology (LIST). We are a multidisciplinary team focused on accelerating the responsible and effective adoption of Artificial Intelligence and Data Technologies across sectors. By combining deep expertise in data analytics, software development, digital twins, AI assessment and evaluation, we support organizations in navigating the complex transition toward scalable, trustworthy, and regulation-compliant digital transformation, with a hands-on approach.

Global context and challenges

The global adoption of Artificial Intelligence (AI) and data-driven technologies is transforming industries, public services, and everyday life. However, this transformation is not without its challenges. Organizations face growing pressure to align AI deployment with evolving legal frameworks (such as the EU AI Act), ethical norms, and sector-specific standards. At the same time, organisations must compose with fragmented data ecosystems, legacy IT infrastructures, and limited capabilities in managing AI life cycles. In addition to governance and compliance, the technological dimension of AI adoption poses substantial hurdles. Issues such as the availability and quality of data, interoperability between systems and in data semantics, explainability of AI decisions, and operationalization of AI models remain open challenges. There is therefore an urgent global need for robust methodologies, validated tools, and testable environments that can guide institutions through the complexity of adopting AI responsibly and effectively.

Mission

The AI Readiness and Assessment group brings together a multidisciplinary team dedicated to accelerating the responsible and effective adoption of Artificial Intelligence (AI) and data technologies across sectors. It supports private and public stakeholders in the adoption of data-driven and AI technologies, by addressing both the technical and compliance aspects of AI readiness with a strong practical approach. We focus on seamless data collection, interoperability and advanced use of predictive and generative AI, while ensuring ethical and legal compliance through rigorous assessments. Our mission is to increase AI readiness by delivering demonstrators and solution prototypes with a high level of application readiness, including advanced data analytics and digital twin technologies. We also aim to provide technical and non-technical guidance, including methodological aspects of AI adoption, encompassing training, capacity building and change management to enable sustainable and impactful digital transformation.

Scope of expertise

This pillar focuses on the foundational elements required for successful AI adoption, with data readiness as a baseline aspect, addressing both the technical infrastructure and the crucial human and organizational factors of AI adoption. Our focus topics include:

Readiness Assessment Models
Data Lakehouses and federated Data Spaces
End-to-End Semantic Integration & Smart Data Models
Privacy-Enhancing Technologies (synthetic data, federated learning)
Efficient Data Pipelines & MLOps

Builds upon Data and AI readiness building tangible, high-impact Smart Cities use cases, with a specific focus on creating descriptive, predictive and prescriptive Local Digital Twins to support Decision-Making in relevant local challenges (e.g., energy, climate, mobility). Our focus topics include:

Local Digital Twin (LDT) Architectures
AI for Energy Transition Scenarios
Predictive & prospective 'What-If' Analysis
Recommender systems for prescriptive twins

Human-in-the-Loop Decision-Support

This pillar addresses the critical need for trust, transparency, and compliance in AI systems, defining AI evaluation frameworks and supporting the development of AI Regulatory Sandboxes with technical components. Our focus topics include:

Legislation-Grounded AI Evaluation Frameworks
AI Regulatory Sandboxes
Bias Detection and Mitigation
AI Validation Labels

Our key assets

Several toolkits for AI assessment, bias testing, and regulatory alignment. Visit the LIST AI Sandbox.

True Local Digital Twins for mobility, energy, and public sector use cases, following EU reference stacks for Local Digital Twins and Data Spaces for Smart Communities.

Platforms for identifying key trend topics in AI, including both skills and adoption metrics.

Several tools for data readiness, predictive analytics and large-scale visualisation.

Experts in AI, Data Engineering, Software Engineering and Regulatory Compliance, with hands-on focus and demonstration-building capacities.

Our collaboration ecosystems

The AIRA Group actively contributes to and benefits from a variety of European and national innovation ecosystems that amplify the impact of our research and technology transfer. We are engaged with:

Testing and Experimentation Facilities (TEFs): As part of collaborative TEFs, AIRA contributes to cross-border and sector-specific AI experimentation platforms that validate emerging solutions in real-world conditions, in particular in the field of Smart Cities
AI Factories: We provide scientific and technical expertise to help companies develop trustworthy AI solutions, create an AI Observatory and Technology Watch, contribute to the Luxembourg AI Sandbox, and implement an "AI Apprenticeship Programme" to upskill mid-career professionals with learning-on-the-job assignments under the supervision of LIST experts. We have a specific focus on the Green Economy sector.
Digital Innovation Hubs (DIHs): We design and deliver services to provide hands-on support to the industrial manufacturing sector in Luxembourg, accelerating AI and data technology uptake via joint training, access to testbeds, and collaborative projects.
Joint Data and AI Labs with Industry: We establish co-innovation spaces with industry partners to pilot and scale AI-based and data-sharing solutions in priority sectors such as finance, mobility, manufacturing, and energy.
European Digital Infrastructure Consortia (EDICs): AIRA engages in the setup and execution of EDICs as a key mechanism for long-term sustainability and governance of strategic digital assets, particularly in the context of Data Spaces for Mobility and Logistics and Local Digital Twins towards the Citiverse.