The Guardian’s Perspective on AI in Warfare: The Iran Conflict Highlights a New Era in Military Strategy

“Never in the future will we move as slowly as we are moving now,” UN Secretary-General António Guterres pointedly remarked this week, underscoring the pressing need to shape the trajectory of artificial intelligence (AI) usage. The rapid pace of technological advancements, coupled with geopolitical upheavals, is blurring the lines between theoretical discussions and tangible events. A political debate surrounding the military’s AI capabilities in the United States has emerged concurrently with the unprecedented usage of AI during the crisis in Iran.

AI firm Anthropic has asserted that it cannot eliminate safeguards that prevent the U.S. Department of Defense from employing its technologies for domestic mass surveillance or fully autonomous weapons of lethal force. The Pentagon has expressed no intentions of pursuing such applications; however, it contends that these crucial decisions should not rest solely in the hands of private companies. Alarmingly, the administration has not only terminated its relationship with Anthropic but also blacklisted the company due to supply-chain concerns. In response, OpenAI has stepped in, maintaining that it has adhered to the limitations established by Anthropic. Yet, in an internal reaction to backlash from users and employees, CEO Sam Altman acknowledged that OpenAI does not control how the Pentagon utilizes its products, expressing regret that the handling of the deal made OpenAI appear “opportunistic and sloppy.”

Nicole van Rooijen, the executive director of Stop Killer Robots—which advocates for retaining human oversight in the use of military force—has warned that the issue transcends the mere possibility of these weapons being utilized; it includes how precursor systems are already altering the landscape of warfare. Van Rooijen cautioned that human control over these systems could easily devolve into an afterthought or become merely a formality.

The shift in paradigm appears to be already underway. Despite the ongoing disputes, reports indicate that Anthropic’s AI model, Claude, has significantly aided in a massive offensive that has claimed the lives of over a thousand civilians in Iran. This has ushered in an era of airstrikes executed “quicker than the speed of thought,” as experts recently told the Guardian. AI systems are now capable of identifying and prioritizing military targets, recommending weapons to use, and assessing the legal justifications for strikes.

It is essential to note that AI technology is not the sole contributor to civilian casualties, military errors, or lack of accountability. U.S. Defense Secretary Pete Hegseth has openly boasted about relaxing the rules of engagement. It is individuals at the Pentagon who are currently evading inquiries regarding the tragic deaths of 165 schoolgirls, resulting from what seems to be a U.S. airstrike on a school in Iran on February 28.

Nonetheless, the adverse impacts of AI are manifest to its operators. An Israeli intelligence source remarked on its application in the Gaza conflict, stating, “The targets never end. You have another 36,000 waiting.” Another intelligence source mentioned spending a mere 20 seconds assessing each target and noted: “I had zero added value as a human, apart from being a stamp of approval.” The ease of mass killing is being amplified significantly, distancing operators from the moral and emotional ramifications of their actions while further eroding accountability.

A pressing call for democratic oversight and multilateral regulations is essential, rather than relegating decisive power solely to corporate and defense entities. As bombs were unleashed over Iran, key global stakeholders convened in Geneva to discuss lethal autonomous weapons systems. During these discussions, a draft text was reviewed that could serve as a solid foundation for a treaty that is urgently required. Most nations are keen to establish clear regulations regarding military applications of AI, yet it is the major players who consistently push back—though they at least participate in the dialogue. The rapid evolution of AI-driven warfare can cast a cautious approach as a liability, potentially handing an advantage to adversaries. However, tech professionals and military officials alike are becoming increasingly aware that the perils of an unchecked expansion far outweigh the risks of controlled advancements.

• Do you have an opinion on the issues raised in this article? If you would like to submit a response of up to 300 words by email to be considered for publication in our letters section, please click here.