Poate AI gestiona autonom arsenalul nuclear global fără drept de veto uman ?

Sistemele actuale de comandă nucleară se bazează deja pe protocoale automate pentru detectare și timpi de răspuns. Avansurile în luarea deciziilor prin AI și integrarea ciber-fizică ar putea curând permite arsenale complet autonome. Riscul de calcul greșit sau escaladare neintenționată crește fără intervenția umană. Strategii militari dezbate dacă astfel de sisteme ar putea fi vreodată de încredere pentru a acționa în cadrul constrângerilor etice. Ar putea supraviețuirea lumii să depindă de mașini care să ia decizia finală în timpul unei crize?

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.