The ACHILLES project is helping organisations navigate the complexities of the EU AI Act in a practical, human-centred way. With full compliance expected by August 2026, this €9 million Horizon Europe initiative is bridging the gap between strict regulations and efficient performance.
Right from the start, ACHILLES has combined robust machine learning practices with a clear focus on sustainability and energy efficiency. Early deliverables—like the D4.1 Legal & Ethical Mapping and D4.4 Ethics Guidelines—translate dense legal documents into straightforward, actionable insights. These outputs have already shaped pilot projects in areas such as healthcare, security, creativity, and quality management.
If you’ve ever struggled to strike the right balance between innovation and regulatory demands, you’ll find the ACHILLES Integrated Development Environment (IDE) a welcome tool. By consolidating regulations, code, and documentation in one place, the IDE makes it easier to stay transparent and compliant, especially during audits.
Experts from diverse fields have welcomed the project. Montaser Awal from IDnow points to ACHILLES as a major step forward in creating privacy-preserving, compliant AI models, while Marco Cuomo and Nuno Gonçalves note its potential to accelerate projects and improve collaborative machine learning—all without sidelining ethics and security.
Guided by the EU AI Act, ACHILLES skilfully integrates norms from GDPR, the Data Act, and the Medical Device Regulation into its framework. A comprehensive legal workshop led by KU Leuven’s CiTiP team has further refined pilot definitions, addressing challenges such as bias and the risk of automation bias along the way.
The project’s methodical compliance loop involves mapping legal norms, designing AI systems that meet these standards, trialling them through pilot tests, and continuously refining the approach. This not only simplifies the development process but also instils confidence in users about AI transparency and safety.
Looking forward, as ACHILLES nears its first-year milestone, further multidisciplinary workshops are on the horizon. These sessions will delve into topics like explainability, bias mitigation, and rigorous compliance verification. In collaboration with initiatives such as ACCOMPLISH, CERTAIN, and DATA-PACT, ACHILLES demonstrates that ethical, sustainable AI is well within reach—and can even provide a competitive edge.
Ultimately, ACHILLES serves as a trusted guide for organisations aiming to blend reliable machine learning with the demands of today’s legal landscape, ensuring that technological advancements and user centricity go hand in hand.
