# How the U.S. Military is Using A.I. to Wage War in Iran Source: https://www.youtube.com/watch?v=x4x1ZsUxgm8 Recap page: https://rapidrecap.app/video/x4x1ZsUxgm8 Generated: 2026-03-13T17:03:34.037+00:00 --- ## Quick Overview The U.S. military is actively using AI, specifically Anthropic's Claude model integrated into Palantir's Maven Smart System, for intelligence processing, mission planning, and suggesting hundreds of targets in the war in Iran, while Iran retaliated by striking cloud data centers in the UAE and Bahrain, highlighting a new front in warfare centered on AI infrastructure. **Key Points:** - The AI model Claude is currently deployed inside classified U.S. military systems for the war in Iran, integrated into Palantir's Maven Smart System to suggest hundreds of targets and prioritize them. - The combination of Maven and Claude has transformed weekslong battle planning into real-time operations, largely using the same model consumers use, potentially with minor fine-tuning for classified environments. - The military currently states humans remain 'in the loop' regarding lethal decisions, but concerns exist that increasing reliance on AI for target selection could lead to scenarios where AI effectively makes the firing decision. - In response to U.S. attacks, Iran struck Amazon data centers in the UAE and damaged one in Bahrain, stating the targets supported enemy military and intelligence activities, demonstrating an attack strategy against AI infrastructure. - The conflict has raised concerns about the security of multi-billion dollar AI infrastructure investments in the Middle East, as data centers are perceived as undefended targets. - A BCG study introduced the concept of "AI brain fry," defined as mental fatigue from excessive use or oversight of AI tools beyond cognitive capacity, experienced by 14% of AI users, distinct from burnout. - Companies like Google and OpenAI quietly removed prohibitions against military use of their AI technologies between 2024 and 2025, reversing earlier ethical stances. **Context:** The discussion centers on the contemporary use of Artificial Intelligence by the U.S. and Israeli militaries in the conflict against Iran, positioning this as a turning point in military technology deployment. Speakers acknowledge the risk of propaganda but confirm notable AI usage, drawing parallels to past technologies deployed abroad that later returned for domestic surveillance applications. Disclosures mentioned include one speaker's fiancé working at Anthropic and the other working for the New York Times, which is suing OpenAI, Perplexity, and Microsoft over copyright. ## Detailed Analysis The analysis confirms the U.S. military is using AI extensively for intelligence gathering, which includes processing massive data troves from sources like hacked traffic cameras and intercepted communications to shrink intelligence 'haystacks' into actionable real-time dashboards, a capability that previously limited manpower prevented. Anthropic's Claude model is cited as the only AI deployed within classified systems so far, integrated into Palantir's Maven Smart System since 2024, where it has suggested hundreds of targets and prioritized them, effectively turning planning into real-time operations. While officials maintain humans are in the loop, experts worry about deferring targeting decisions to increasingly capable AI systems, referencing a missile strike on an Iranian school that killed civilians as a potential, though unconfirmed, AI-related incident. Conversely, Iran targeted cloud infrastructure, attacking Amazon data centers in the UAE and Bahrain to disrupt services supporting allied military activities, signaling that data centers and fiber optic cables are now critical battlefronts. Separately, research reveals negative psychological effects on workers, with BCG identifying "AI brain fry"—cognitive strain from excessive oversight—as distinct from burnout; this fatigue is highest in rapidly adopting fields like marketing due to increased iteration and oversight demands, prompting calls for redefining AI fluency to include cognitive health. ### Military AI Deployment in Iran - Israeli intelligence previously monitored Tehran traffic cameras and eavesdropped on officials - AI excels at processing large data quantities from drones and sensors onto real-time dashboards - Current use focuses on intelligence, mission planning, and logistics, stopping short of fully autonomous lethal weapons. ### Role of Claude and Maven - Claude is the only AI model deployed in classified systems in the Iran conflict - Maven Smart System, built by Palantir and integrating Claude since 2024, suggests hundreds of targets and issues precise location coordinates - This combination turns weekslong battle planning into real-time operations. ### Ethical and Corporate Shifts - Pentagon declared Anthropic a supply chain risk amid conflict - Google quietly removed its AI principle prohibiting military use in 2025, as did OpenAI in 2024 - Anthropic is noted as the only frontier lab that never explicitly prohibited military applications. ### Iran's Counter-Strategy - Iran struck Amazon data centers in the UAE and damaged one in Bahrain after initial U.S. attacks - Iran explicitly targeted data centers to disrupt enemy military and intelligence support activities - This action prompts questioning of placing multi-billion dollar AI infrastructure in the Middle East due to perpetual risk. ### AI's Impact on White-Collar Workers - BCG study found 14% of AI users experience "AI brain fry," defined as mental fatigue from excessive oversight beyond cognitive ability - Brain fry correlates with increased mental effort and information overload, particularly in marketing roles - Using AI for repetitive tasks, however, was found to reduce burnout and increase social connection. ### Historical Parallels and Future Work - Historical parallels drawn to the 1970s Lordstown syndrome regarding worker alienation from automation - Researchers suggest redesigning work and focusing on outcomes over output is necessary to mitigate brain fry - Workers are advised to engage managers and teams to integrate AI workflows better, though fears of job loss impede open discussion.