IDnow to create synthetic data for EU-funded AI ethics project

Identity verification platform IDnow has officially joined the European Union-funded ACHILLES project, an initiative addressing the ethical and technical challenges of artificial intelligence (AI).

With 16 partners and 8 million euros (US$.8.2 million) in funding from the Horizon Europe program, ACHILLES is designed to tackle the “Achilles heel” of AI – efficiency and trust. The goal is to create systems that are not only more effective and efficient but also ethically responsible, transparent and compliant with European regulations, according to the firm’s release.

The initiative will test real-world world applications in industries such as healthcare, pharmaceuticals and identity verification.

IDnow will provide its knowledge in ID verification to increase the efficiency of datasets while ensuring they comply with privacy rules such as the GDPR. To achieve this, the company will tackle creating synthetic data that can replace real data.

The biometrics industry has been relying on synthetic data to tackle issues such as the lack of data, bias and data privacy restrictions. The EU’s AI Act may make the gathering and processing of biometric data even more complicated for identification firms and synthetic data such as faces created by generative AI could solve a part of the problem.

“Thanks to the ACHILLES project, we hope not only to improve the quality and robustness of our algorithms but also to reduce dependence on real data,” says Montaser Awal, Director of Research at IDnow. “This represents a major step toward more sustainable and regulatory-compliant AI.”

The Germany-based firm will also work on optimizing data annotation to minimize human labor and renewing datasets without affecting model performance.

The project will run over four years, ending in 2028. It counts six objectives, including identifying ethical and legal frameworks for designing AI systems, creating frameworks for quality data for AI, consumer protection, reduce energy and IT costs during AI model training and creating robust and transparent algorithms.

“Our aim is to ensure that every technology not only performs well but is also transparent and ethical, to inspire trust in the long term,” says André Carreiro, the project’s coordinator and a researcher at the applied research organization Fraunhofer AICOS.

Aside from IDnow, ACHILLES also includes research institutions, universities, healthcare organizations and technology providers. Among the partners are the KU Leuven Center for IT & IP Law, Fraunhofer, Institute for Systems and Robotics (ISR) and the Institute of Systems and Computer Engineering – Research and Development (INESC-ID) in Lisbon, the Portuguese Mint and Official Printing Office (INCM) and more.

Article Topics

ACHILLES project  |  digital identity  |  ethics  |  EU  |  IDnow  |  research and development  |  responsible AI  |  synthetic data