Trustworthiness For AI in Defence: Developing Responsible, Ethical, and Trustworthy AI Systems for European Defence

The purpose of this document is to collect,present and describe the aspects of Trustworthiness for AI in Defence in a‘food for thought’ approach reflecting the combined view of AI experts andstakeholders from Defence Industry, Academia and Ministries of Defence. This effort is performed in the context of the European Defence Agency’s (EDA) Action Plan on Artificial Intelligence for Defence and tries to address thetopics of trusted AI and verification, validation and certificationrequirements analysis. The topics covered and analysed in this document willprovide the appropriate knowledge of the current global status considering the AI regulations, standards and frameworks for AI trustworthiness and will alsorecommend the follow-up activities that will further assist the EU MembersStates and Defence Industry to better prepare, plan and develop the future AIsystems aligned with the identified expectations. The target audience is EU Member States especially the MODs that will furtherevaluate the whitepaper’s recommendations and may use it as a reference pointfor future related AI research activities. Target audience will be expandedalso to Defence Industry to highlight the key aspects and requirements fordeveloping AI systems for military use considering all regulations, standardsand methodologies assisting them in better planning and design to delivertrusted AI services and applications. The report is structured based on theidentified tasks as follows: • AI definition and Taxonomy • Identification of stakeholders of AI indefence • Legal perspective for AI use cases andscenarios • Standards and Regulations • Testing & Evaluation, Validation andVerification standards, tools and methodologies • Human Factors in Defence • Ethical AI considerations • Impact analysis of AI in Defence • Military Use cases and scenarios for Trustworthiness of AI • Recommendations and next steps. This whitepaper is the starting point andan initial reference source for future related AI research activities andproject proposals that essentially will provide to the community the adequateinformation and knowledge of how to plan, develop, acquire, test and usedefence AI Systems. The whitepaper is also affected by therecently approved EU AI Act which is a regulation that is expected to havea significant impact on the future design and use of AI technology across EU. As part of the latest updates regarding theAI Act’s progress, the Commission issued a standardisation request, taskingCEN and CENELEC with developing European standards by 30 April 2025. These standards aim toensure AI systems in the EU market are safe, uphold fundamental rights, andencourage innovation. Either way the recommendations of thisdocument, without raising any commitment or obligation from the Member States,are expected to feed partially the proposed AI Action Plan update (v.2.0) that will further address inmore detail the topics raised both in this paper but also any other identifiedtopic that will be introduced and accepted over the action plan’s developmentprocess. It must be noted that this whitepaper ismainly an outcome of a work done in a volunteer basis by the Trustworthinessfor AI in Defence Working Group with multidisciplinary expert members fromIndustry and Academia across Europe but also members from some EU MODs. Licence: CC BY-NC-ND