The initiative will develop cyber risk evaluation tools and compliance methods to support enforcement of the EU AI Act.
Capgemini has been selected, as leader of a consortium[1], by the European Commission’s AI Office[2], established within DG CNECT (Directorate-General for Communications Networks, Content and Technology), to support the cyber threat modeling and risk assessment related to advanced AI systems, in the context of the European Union’s AI Act. The assignment is part of the EU’s AI Safety[3] program and focuses on assessing and mitigating cyber offence risks associated with General‑Purpose AI (GPAI) models[4].
Capgemini will help strengthen Europe’s ability to assess, monitor and mitigate the systemic cyber offense risks associated with advanced General Purpose AI (GPAI) models. As these models rapidly gain autonomous and cyber-relevant capabilities, such as automated vulnerability discovery, exploit generation, or large-scale cyber-attack orchestration, Europe needs robust harmonized evaluation frameworks to ensure these technologies are deployed safely and responsibly across the EU.
Capgemini brings its expertise in AI, cybersecurity and compliance to support the AI Office in its core mission: the development and use of trustworthy AI. This includes modelling emerging cyber offence risks, developing evaluation tools, designing technical compliance methodologies, and conducting on‑demand technical assessments. Capgemini will support the AI Office to identify systemic risks and effectively enforce the AI Act’s obligations for GPAI providers.
Capgemini is leading a dedicated consortium with Wavestone and the French public institution LNE (Laboratoire National de Métrologie et d’Essais), bringing together complementary expertise in cybersecurity risk management, operational security, AI evaluation and conformity assessment to help the European Commission keep pace with fast-evolving AI-driven cyber threats.
For more information, visit capgemini.com.
[1] Under Lot 2 of the EC-CNECT-2025-OP-0032 procurement procedure.
[2] The European AI Office is the centre of AI expertise across the EU, that promotes the development and deployment of AI solutions that benefit society and the economy.
[3] AI Safety Program consists in developing tools, methodologies and benchmarks for evaluating capabilities and reach of general-purpose AI models and classifying models with systemic risks. See https://digital-strategy.ec.europa.eu/en/policies/ai-office
[4] https://artificialintelligenceact.eu/gpai-guidelines-overview/
