The study, entitled “Ethical Considerations for the Military Use of Artificial Intelligence in Visual Reconnaissance”, discusses ethical issues and principles that must be considered when using AI in military scenarios. A particular focus is on the so-called FATE guidelines, which include fairness, accountability, transparency and ethics. These established guidelines are intended to ensure that AI-supported systems are used transparently, fairly and responsibly in the military sector.
One contribution of the study is to incorporate European and German values into the discussion of ethical principles for military AI applications. In addition to established principles such as traceability, responsibility and reliability, these also include principles such as proportionality and manageability. The ethical principles are examined in three realistic use cases from the air, land and sea. Methods such as automated sensor data analysis and eXplainable AI (XAI) are used to put the scenarios in the direct context of HENSOLDT's product portfolio and technical feasibility. The study concludes that ethical principles can be taken into account when using AI in a military environment. This applies, for example, to technologies such as XAI or human-machine interfaces based on Large Language Models (LLMs). Ultimately, responsibility for a decision remains with the human being who retains control over the AI.
“Our research represents an important step in the discussion on how to use artificial intelligence responsibly in the complex and dynamic landscape of military scenarios,” said Dr Michael Teutsch, AI Strategist & Lead at HENSOLDT. “It shows how we as a company can help to align technological advances with ethical standards, striking a balance between the benefits of AI and compliance with moral and legal requirements.”
The study is available here.
