The U.S. Airstrike on Iran's Defense Technology Marvel: AI Warfare, LUCAS Drones on the Battlefield

The Wall Street Journal reports that during the U.S. airstrikes on Iran, the U.S. Central Command used Anthropic’s Claude model for intelligence assessment, target analysis, and operational scenario simulation. On the same day, the Trump administration announced that Anthropic was designated as a “national security supply chain risk.” This article introduces the new defense technology leaders on the battlefield that span AI and drones.

Claude Blacklisted but Appears in Airstrikes

Anthropic is the first U.S. company to deploy AI models on the Department of Defense’s classified networks, but disagreements over usage terms have escalated. The Pentagon demands that Claude be accessible for all lawful military purposes. Anthropic insists on two red lines: it cannot be used for mass surveillance of U.S. citizens or for fully autonomous weapons systems.

Anthropic CEO Dario Amodei publicly stated before the deadline, “We cannot in good conscience agree to their demands.” He emphasized that while the AI company understands that military decision-making is under government control, in certain specific scenarios, AI might undermine rather than defend democratic values.

(Anthropic clashes with the Pentagon! CEO Amodei refuses “domestic mass surveillance” and “fully autonomous weapons”)

Defense Secretary Pete Hegseth announced that Anthropic is a “supply chain risk,” a label usually reserved for foreign adversaries, and ordered all military contractors to immediately cease business with Anthropic. Simultaneously, Trump ordered federal agencies to stop using all Anthropic technology. On the same day the ban was announced, media reports indicated that U.S. Central Command was still utilizing Claude during the Iran airstrike operations for intelligence analysis and target assessment.

OpenAI Capitalizes on the Opportunity to Take Over the Pentagon

A few hours after the ban, OpenAI CEO Sam Altman announced that they had finalized an agreement with the Pentagon, granting OpenAI tools access to classified U.S. military systems, with usage restrictions nearly identical to those originally requested by Anthropic. One of the biggest winners in this conflict is OpenAI.

U.S. Airstrikes on Iran, LUCAS Drones Enter Combat

In this epic operation, the U.S. also announced the deployment of the LUCAS drone (Low-cost Unmanned Combat Attack System). Each LUCAS unit costs only $35,000, offering a significant cost advantage and easier large-scale deployment. The Department of Defense has indicated it will adopt a free-for-all procurement model, contracting multiple vendors rather than relying on a single supplier.

Palantir: From Maduro to Iran, AI Combat Systems Fully Operational

Palantir has already established a core position in this defense AI race. In January, U.S. forces used Palantir’s platform combined with the Claude model to arrest Venezuelan President Maduro, one of the early examples of AI-assisted real military missions.

Palantir currently holds an astonishing scale of defense contracts: in July 2025, it secured a $10 billion enterprise service agreement with the U.S. Army, consolidating 75 existing contracts into a single framework covering data integration, analysis, and AI tool deployment.

Its flagship product, TITAN (Tactical Intelligence Targeting Access Node), is the most prominent AI combat platform. TITAN is a mobile ground station integrating real-time data streams from satellites, drones, and sensors, using AI algorithms to automatically identify threats and calculate strike coordinates, enabling soldiers to complete missions more quickly.

Ondas: Fully Automated Drone Swarms Enter European Airports

Compared to Palantir’s data analysis and command positioning, Ondas Holdings focuses on highly automated tactical execution. Its core product, Iron Drone Raider, is an autonomous anti-drone system that continuously monitors airspace with multiple sensors and AI algorithms. When an intruding drone is detected, it autonomously dispatches interceptors that capture targets with reusable net payloads, aiming to reduce operator intervention and avoid interference with civilian communications and infrastructure.

By late 2025, Ondas announced two major European airport defense orders worth approximately $8.2 million each, making Iron Drone Raider a key solution for sensitive areas like airports. Border and critical infrastructure defense are viewed by the company as the most promising future deployment scenarios.

Ondas’ Autonomous Systems division (OAS) is involved in multiple government and institutional projects, aiming to deploy large-scale autonomous drone swarms at fixed and mobile sites for persistent ISR, rapid response, and automated threat suppression. The overall architecture emphasizes integrated command and control of aerial and ground platforms, combining AI classification and electronic warfare capabilities to enhance border and critical infrastructure security and response.

Anduril Sanctioned by China for Delivering Drones to Taiwan

Founded in 2017, Anduril specializes in AI and automation systems, reshaping traditional defense R&D and procurement with AI, autonomous systems, and sensor networks. Its products include Ghost drones, Anvil anti-drone defense systems, base and border surveillance towers, and the Lattice command platform that integrates battlefield data. Major clients include the U.S. Department of Defense and allied nations.

Compared to traditional defense giants like Boeing and Lockheed Martin, Anduril is known for rapid iteration, vertical integration, and extensive use of commercial technology, representing a new generation of defense industry companies.

Last year, Anduril founder Palmer Luckey visited Taiwan, delivered the Altius-600M attack drone, met with Defense Minister Guo Lihue, and gave a public speech at National Taiwan University, discussing Taiwan’s security challenges and future defense strategies. He and his company were added to China’s sanctions blacklist for selling military equipment to Taiwan. He remains relaxed, saying it’s like receiving a Christmas gift.

(Delivered drones to Taiwan, sanctioned by China! Anduril founder Palmer: like receiving a Christmas gift)

Shield AI Collaborates with NCSIST to Introduce Domestic AI Flight Control into Drones

Another defense startup, Shield AI, founded by 曾國光, who comes from a Taiwanese diplomatic family in the U.S. and served in the U.S. Navy SEALs. Shield AI has signed a cooperation agreement with Taiwan’s National Chung-Shan Institute of Science and Technology (NCSIST) to integrate its core AI platform Hivemind into Taiwan’s self-developed unmanned systems, accelerating the development, testing, and deployment of domestic AI flight control and autonomous combat capabilities. 曾國光 also pledged to expand investments in Taiwan over the next five years.

(US defense startup Shield AI partners with NCSIST to develop drones! Founder 曾國光’s Taiwan roots revealed)

According to the agreement, NCSIST will utilize the Hivemind platform to accelerate the development of local AI flight control systems capable of autonomous missions and multi-unit cooperative combat, integrating them into Taiwan’s various unmanned platforms. In the future, a single ground control station will allow one operator to monitor and command multiple drones simultaneously, reducing manpower and increasing operational efficiency. Shield AI has also established an office in Taipei 101 and is actively recruiting local high-tech talent, committed to long-term investment in Taiwan’s defense and industrial ecosystem.

This article, “The Hidden Defense Tech Pioneers Behind U.S. Airstrikes on Iran: AI Warfare, LUCAS Drones on the Battlefield,” originally appeared on ABMedia.