- What: Palantir integrated Anthropic’s Claude AI into military software to generate attack plans and nominate targets.
- Tools: The "Maven Smart System" and "AI Asset Tasking Recommender" use AI to suggest munitions and bomber assignments.
- Conflict: Anthropic has filed two lawsuits against the U.S. government after the Pentagon labeled its AI a “supply-chain risk.”
- Operations: Claude has reportedly been utilized in U.S. military operations in Iran and the capture of Venezuelan President Nicolás Maduro.
Data analytics giant Palantir and AI startup Anthropic have demonstrated how generative AI chatbots are now being used by the U.S. military to synthesize intelligence, order drone reconnaissance, and generate complex war plans. The integration, which began in November 2024, allows military analysts to use a ChatGPT-style interface to "nominate" targets for bombardment and organize the jamming of enemy communications in real-time.
The Dawn of Algorithmic Warfare
Recent software demonstrations and public records reviewed by WIRED reveal a significant shift in how the Department of Defense (DOD) utilizes large language models (LLMs). While AI has long been used for data sorting, Palantir’s latest tools—including the "Maven Smart System"—apply computer vision algorithms to satellite imagery to automatically detect "enemy systems."
According to a demo presented by Cameron Stanley, the Pentagon’s chief digital and artificial intelligence officer, the system can distinguish between civilians and military vehicles with high precision. Once a target is identified, a tool known as the "AI Asset Tasking Recommender" suggests which specific bombers and munitions are best suited for the strike. This level of automation marks a transition from AI as a back-office tool to a core component of kinetic military operations.
The Anthropic-Pentagon Legal Battle
Despite the rapid adoption of these tools, a "heated dispute" has erupted between the Pentagon and Anthropic. In February, Anthropic reportedly refused to grant the U.S. government unconditional access to its Claude models, citing concerns over mass surveillance and the development of fully autonomous weapons. In response, the Pentagon designated Anthropic’s products as a “supply-chain risk.”
Anthropic struck back this week, filing two lawsuits alleging illegal retaliation by the Trump administration. The startup is seeking to overturn the "supply-chain risk" designation, which could jeopardize its lucrative partnership with Palantir and its standing within the defense industry. While both companies declined to comment on the litigation, the clash underscores the growing tension between Silicon Valley’s ethical guidelines and the military’s operational requirements.
From Satellite Data to "Target Intelligence"
Palantir’s primary vehicle for AI deployment is "Project Maven," also known as the Algorithmic Warfare Cross-Functional Team. Managed by the National Geospatial-Intelligence Agency (NGA), Maven is being deployed across the Army, Air Force, Space Force, Navy, and Marine Corps.
The software integrates multiple data streams into the Army Intelligence Data Platform (AIDP), which creates "intelligence running estimates." These are live, frequently updated collections of battlefield information that precede final intelligence summaries. While Palantir has confirmed that Claude helps analysts uncover "data-driven insights," it has remained vague on exactly which systems rely on the chatbot. However, reports from the New York Times and Washington Post indicate that Maven’s core functionality is now heavily dependent on Anthropic’s technology.
Impact on Global Conflict and Strategy
The real-world implications of these AI integrations are already being felt. Claude reportedly played a critical role in the January U.S. military operation that led to the capture of Venezuelan President Nicolás Maduro. Furthermore, the AI tool is currently being used by U.S. Central Command to oversee escalating military operations in Iran.
For developers and military strategists, this represents a new era of "decision dominance." The ability to process vast amounts of intelligence and generate multiple attack plans in seconds could provide a decisive edge in "time-sensitive situations." However, the lack of transparency regarding how Claude handles sensitive queries or provides recommendations raises significant concerns about accountability in AI-assisted warfare.
"This changes how commanders visualize the battlefield; for the first time ever, a chatbot is recommending who to strike and what munition to use," according to details from recent military assessments.
What’s Next for AI in the Military
The outcome of Anthropic’s lawsuits against the government will likely set a major precedent for how AI companies interact with the defense sector. If the "supply-chain risk" designation stands, it could force Palantir to seek alternative LLM providers for its Maven and AIDP platforms, potentially slowing the rollout of generative AI in combat zones.
In the near term, Palantir CEO Alex Karp has signaled that the company will continue to deepen its ties with Middle East partners, using Palantir’s platform to coordinate data between the U.S. and allies targeted by Iranian airstrikes. As Maven continues its deployment "across the entire department," the line between human decision-making and algorithmic execution continues to thin.