Grok AI and the new era of AI-enabled warfare: US drone targeting and operations in Iran and Ukraine

The US Department of Defense (often referred to in this context as the Department of War) has cast xAI's data centers as battlefield infrastructure of "paramount national security" importance. Elon Musk's Grok AI has become an integral part of America's war machine, according to a sworn declaration by the Pentagon's artificial intelligence chief. Its data centers must be treated with the same priority as munitions production plants in future conflicts.

The statement about Grok was submitted as part of an effort by the Trump administration to intervene in a lawsuit seeking to restrict the operation of xAI's Colossus data centers in Memphis, Tennessee (and associated facilities), over environmental concerns related to gas turbines powering the supercomputer cluster.

Cameron Stanley, the chief digital and artificial intelligence officer at the Department of Defense, compared xAI's computing capacity to weapons manufacturing. He argued that the company's ability to operate data facilities "at massive scale" is "as foundational to our modern defense posture as traditional munitions production."

Stanley revealed that Grok is a crucial part of Palantir-developed Maven Smart Systems (MSS), a platform supporting "targeting, intelligence, readiness, and recruitment," as well as military planning and logistics. "MSS frontier workflows enabled US forces to deploy over 2,000 munitions to 2,000 distinct targets within 96 hours during Operation Epic Fury," he said. xAI is one of just a few enterprise providers able to sustain "mission-critical operations" across Secret and Top-Secret classified networks. He warned that disruptions to xAI's Colossus facilities could impair Pentagon capabilities.

AI in drone targeting and navigation: Iran campaign

The reference to Maven places Grok inside a broader and increasingly controversial AI targeting ecosystem. Project Maven, originally an algorithmic warfare initiative to process vast battlefield data from drones, satellites, and sensors, has evolved significantly. Palantir has marketed its software as an "AI-powered kill chain" promising "decision dominance."

In Operation Epic Fury, the major US-led strikes against Iran, AI systems, including integrations with models like Grok and Anthropic's Claude via Maven, accelerated target identification, prioritization, and strike coordination. This enabled rapid strikes on thousands of targets in the opening phases, compressing what once took days into hours.

Drones played a central role, with the US deploying systems like the Low-Cost Uncrewed Combat Attack System (LUCAS), a one-way attack drone inspired by Iranian Shahed designs. AI supports these through enhanced navigation in jammed environments, swarm coordination, object detection/classification via computer vision, and real-time decision support when human control is limited. Grok and similar frontier models contribute to intelligence fusion, logistics, natural language querying of data, and workflow automation in these systems.

However, this speed came with costs. Reports linked AI-assisted targeting (via Palantir-linked systems incorporating large language models) to a strike in Minab that killed nearly 160–180 young schoolgirls at the Shajareh Tayyebeh primary school. The school was reportedly on an outdated target list, possibly misidentified as a military site adjacent to an IRGC complex. Anthropic CEO Dario Amodei noted that the use did not necessarily violate the company's "red lines," as a human made the final decision. Critics argue that AI-driven acceleration can amplify errors from stale data or insufficient human review.

Palantir executives have embraced the military role. CEO Alex Karp has described the company's purpose as "to scare enemies, and on occasion, kill them," while its manifesto highlights "hard power in this century will be built on software" and AI-enabled deterrence.

Role in Ukraine and broader drone warfare lessons

AI's contributions extend beyond Iran. In the Russia-Ukraine conflict, both sides have pioneered low-cost drone swarms, FPV (first-person view) kamikaze drones, and autonomous navigation under heavy electronic warfare (jamming). The US has drawn lessons from Ukraine's battle-tested counter-drone technologies and interceptor drones, deploying similar systems (some developed with Ukraine in mind) against Iranian threats.

Grok and xAI's compute resources support broader US AI efforts, including data processing for drone autonomy, target recognition, and logistics in contested environments. Ukraine's experience with AI for real-time threat detection and swarm tactics has influenced US doctrine, with Maven-style platforms helping integrate multi-domain data (air, sea, ground, cyber) for drone operations. Reports indicate AI compresses the OODA loop (Observe-Orient-Decide-Act), enabling faster responses in drone-heavy battlefields.

xAI, which developed Grok as a rival to models like ChatGPT, has seen its infrastructure merged or closely tied with SpaceX efforts. Colossus provides compute not only for Grok but rented to others, including Google and Anthropic, amplifying its strategic value.