The paper presents theoretical considerations regarding the understanding of strategic autonomy in the field of security and defence. It starts with the theoretical understanding of the term “autonomy” and dilemmas concerning autonomy in the EU. Then it identifies and describes the key initiatives in the field of security and defence conditioning the EU’s achievement of strategic autonomy in this area and the main problems of their implementation. The presented conclusions are based on the qualitative analysis of the source material, mainly, the EU normative documents. They lead to the following observations. First, there is no clear definition and interpretation of “strategic autonomy” in the EU normative documents. This can lead to confusion and over-interpretation by individual Member States which may understand strategic autonomy differently, especially in the area of security and defence. Secondly, the majority of the Member States recognise security and defence as an area enabling the achievement of strategic autonomy. However, there are differences between countries in terms of understanding strategic autonomy. Two approaches are visible amongst the EU members: full sovereignty and flexible autonomy in the field of security and defence. Thirdly, the security and defence initiatives adopted by the EU over the past few years can provide the basis for achieving strategic autonomy in this area.
Advanced studies on artificial intelligence increase the concerns of many non-governmental organizations regarding the possibility of their use in combat systems. Even now, there are many weapon systems that can interact with the enemy in an automated manner. However, the concerns are raised due to the potential emergence of autonomous combat systems, which will independently make decisions about life or death on the battlefield, both in relation to military personnel and civilians. This article provides a review of terminological issues associated with the concept of automation and autonomy in the context of combat systems. It was concluded that the absence of a commonly accepted definition of “lethal autonomous weapon system” within international law contributes to many interpretations in this field. On the basis of source literature, the second part of this article presents current proposals aimed at a preventive ban on the design and use of lethal autonomous weapons systems. The essential deduction leads to the conclusion that introduction of a total ban on the design and use of this type of weapon systems is impossible, nevertheless as much as possible should be done in order to make sure that the critical decisions, especially those about life and death, will remain in the scope of human control and that they will not be made by the machines.