Artificial intelligence is poised to revolutionize nearly every facet of warfare, from logistics and planning to cyber operations and battlefield targeting. AI is being used to optimize military supply chains and enhance efficiency in moving resources globally. In cyber warfare, AI is employed for both offensive actions, such as implementing cyberattacks, and defensive measures, including analyzing foreign intelligence. The most contentious application of AI on the battlefield involves high-level military decision-making, particularly in identifying and recommending attack targets, raising significant ethical concerns about potential autonomous decision-making by AI.
The prospect of AI systems independently making targeting decisions, often referred to as "pushing the button," is a major source of anxiety, though it is not yet fully realized. However, military leaders are actively planning for this future. The discussion around an AI arms race, particularly between major players like the US and China, is ongoing. Historically, military technological leadership was linked to high budgets and expensive, sophisticated equipment. However, the conflict in Ukraine has demonstrated a shift, highlighting the effectiveness and expendability of cheaper, AI-enabled drones, thereby changing the dynamics of who is considered ahead in military innovation.
Countries like the UK are investing heavily in AI for military applications, with $1 billion earmarked for a "digital targeting web" to integrate data from various sources and enable faster, more informed command decisions. The US has already deployed AI-powered command and control systems, such as Maven, reportedly used in combat in Yemen in 2024, with NATO adopting similar variants. Ukraine's "drone war" is accelerating the development of autonomous weapons, with nearly 2 million drones acquired in 2024, of which 10,000 were AI-enabled. Companies like Helsing, valued at €4.95 billion in 2024, are supplying thousands of AI-enabled attack drones to Ukraine. This rapid advancement raises serious ethical questions about fully autonomous weapons and the challenges of regulation given the speed of technological development and the necessities of warfare.
While full autonomy remains some years away, AI is already enhancing human decision-making under pressure. Current military uses of AI include planning, logistics, cyber warfare, and controversial weapons targeting, which is already in use in Ukraine and Gaza. For instance, Kyiv uses AI to guide drones past jammers, and the Israel Defense Forces developed Lavender, an AI-assisted system that identified approximately 37,000 potential human targets in Gaza. There has been a notable shift in attitude among AI companies, with some, like OpenAI, initially forbidding military use but later partnering with defense tech firms due to factors like the allure of advanced, more precise warfare, the need to recoup significant R&D investments, and the substantial funding available from military budgets and venture capitalists in the defense tech sector.
Criticisms of AI in warfare include concerns that increased precision might lead to more frequent conflicts rather than fewer casualties, drawing parallels to the impact of drone warfare in Afghanistan. Experts like Missy Cummings, a former US Navy fighter pilot and engineering professor, voice skepticism about the reliability of large language models in high-stakes military scenarios, cautioning against the "extraordinarily big promises" made by some tech companies. The consensus is that while AI offers transformative capabilities for warfare, critical questions about safety, oversight, and ethical implications demand continued scrutiny and accountability from political leaders.
