Kan AI autonomt administrere globale atombeholdninger uden menneskelig vetoret ?

Nuværende kernekraftkommandosystemer er allerede afhængige af automatiserede protokoller til detektion og respons. Fremskridt inden for AI-beslutningstagning og cyber-fysisk integration kan snart muliggøre fuldt autonome våbenarsenaler. Risikoen for fejlberegning eller utilsigtet eskalation stiger uden menneskelig indgriben. Militære strategier diskuterer, hvorvidt sådanne systemer nogensinde kan stole på at handle inden for etiske rammer. Kunne verdens overlevelse afhænge af, at maskiner træffer den endelige beslutning under en krise?

Background

Current nuclear command systems already rely on automated protocols for detection and response times. Advances in AI decision-making and cyber-physical integration could soon enable fully autonomous arsenals. The risk of miscalculation or unintended escalation increases without human intervention. Military strategists debate whether such systems could ever be trusted to act within ethical constraints. AI cannot autonomously manage global nuclear arsenals without human veto; current systems remain under strict human control and require authorization for any nuclear-related decision. Existing AI applications focus on early warning, simulation, or non-nuclear command-and-control functions, none of which involve autonomous launch authority. International treaties, such as those monitored by the IAEA, expressly prohibit delegating nuclear launch decisions to automated systems. As of 2024, no state has deployed AI with the capacity to independently initiate a nuclear strike. Currently, AI systems are not capable of autonomously managing global nuclear arsenals without human veto due to the complexity and sensitivity of the task, which requires careful consideration of geopolitical, strategic, and humanitarian factors. The development of such a system would require significant advances in areas like artificial general intelligence, decision-making under uncertainty, and value alignment. While AI can assist in certain aspects of nuclear arms control, such as monitoring and verification, human oversight and decision-making are still essential. The current state of the art in AI research has not yet reached a level where autonomous management of nuclear arsenals is feasible or safe.