In February 2026, during the launch of the illegal and unauthorized war with Iran, the United States attacked a girl’s school in the city of Minab, killing at least 168 people. Since the travesty, more than 160 members of Congress across the House of Representatives and Senate, have demanded answers regarding what caused this attack to occur. While the official investigation is ongoing, early reporting has indicated that “overreliance on artificial intelligence (AI) systems” may be partly to blame.
This tragedy is just one of the many reasons FCNL is increasing our focus on challenging the Pentagon’s increasing use of artificial intelligence, especially in the targeting and control of deadly weapons systems. We have joined the Stop Killer Robots campaign, an international coalition advocating for the development of legal frameworks and other norms that ensure meaningful human control over the use of force. We are lobbying Congress to pass responsible regulations and guardrails.
“To disarm [AI] does not mean rejecting technology but preventing it from dominating humanity.”
- Pope Leo XIV, “Magnifica Humanitas”
The Pentagon’s Ongoing War on AI Guardrails
Over the past year, the Pentagon has been locked in a legal battle with the AI company Anthropic. The company publicly opposed the Pentagon’s plan to use AI for domestic surveillance and fully autonomous weapons, ultimately causing them to lose their contract. Recent reporting has indicated that the Pentagon wanted its own AI technology, which would rarely say no to military orders.
The Minab school strike and the legal battle with Anthropic have not slowed the Pentagon’s push for this technology. Secretary of Defense Pete Hegseth has recently announced a new Pentagon command office called “Autonomous Warfare Command,” which would focus on building and providing autonomous weapons across the U.S military. Hegseth says the priority is speed, highlighting the dangerous arms race narrative against other countries like China as the political argument being used to push investments in AI and autonomous weapons forward. Meanwhile, a new report by the U.S Government Accountability Office (GAO) states that the ongoing AI projects at the Pentagon are not consistently following existing ethics guidelines.
This raises serious questions: When an AI system is partly to blame for civilian casualties or war crimes, who is held responsible? How can Congress ensure this technology will be used responsibly if the Pentagon is already choosing to ignore existing ethics rules?
This “accountability gap” is just one of the problems Congress needs to urgently address, especially as the Pentagon continues to invest heavily in this technology.
From Anxiety to Action
In 2025, FCNL updated its policy statement to reflect that,
“It is essential that federal legislation and regulatory frameworks ensure AI applications are ethical and protect the public, especially children and other vulnerable populations, from AI misuse… We oppose autonomous AI weapons.”
FCNL has since supported legislation and executive actions aimed at requiring a “human in the loop” as it relates to nuclear weapons, primarily focused on ensuring AI can never independently launch or detonate a nuclear weapon on its own. We also included recommendations aimed at improving AI guardrails in this year’s National Defense Authorization Act (NDAA), including provisions which would ensure a human is always in charge and accountable when these systems use force, require more testing and safety checks on AI weapons, and set up a system for reporting when AI goes wrong.
This is only the beginning. If trends continue, this technology is poised to become an increasingly larger part of future wars. FCNL is engaging with a growing number of partners to better understand the risks posed by AI and press for legislative safeguards. We expect this to be a bigger part of our legislative work on militarism with the incoming 120th Congress. Congress cannot continue to allow the Pentagon, or private companies, to set up the rules for this new frontier of technology.
It is not too late to stop this new AI arms race mentality from getting out of hand. As we reflect and look ahead, we recommit ourselves to the important work of protecting civilian populations from conflict, including the risks of emerging technology associated with weapons of war.
