Army’s CamoGPT won’t be phased out as Pentagon embraces more commercial genAI products
In the aftermath of the Pentagon’s large-scale rollout of the new GenAI.mil platform, the Army has no plans to deactivate its highly specialized CamoGPT machine learning offering or cut off personnel’s access to the other preexisting, purpose-built frontier AI models.
“The U.S. Army remains committed to advancing artificial intelligence capabilities through rigorous research and development,” a spokesperson for the service told DefenseScoop this week.
Generative AI is an emerging and disruptive technology field that applies massive models to generate original, human-like outputs that are based on user prompts, but not always factual. Since commercial genAI products were widely released to the general public around late 2022 — quickly becoming a global phenomenon — experts have warned about severe known and unpredictable threats that the tech poses to humanity, public trust and democratic processes.
Broadly, the individual U.S. military services’ policies and paths to adopting genAI systems have been fragmented and complicated by data ownership, acquisition issues and other complexities.
The Army’s AI Integration Center (AI2C) began prototyping CamoGPT in 2024. That model is billed as more than a chatbot, and is designed to reliably support personnel across a range of administrative and operational tasks in a safe environment. Thousands of users reportedly interact with the tool daily to receive insights built off Army-specific data, like internal doctrine and training materials, while maintaining security for Controlled Unclassified Information (CUI) and certain classified information.
Building on that work, in 2025, the Army unveiled its Enterprise Large Language Model Workspace, which the spokesperson said currently provides insiders with access to more than 23 approved, industry-made AI frontier models.
Separately, Pentagon leaders led the breakneck launch of GenAI.mil last month.
Through that platform, the Defense Department aims to supply more than three million military personnel, civilian employees and contractors with access to advanced, commercial-grade genAI models. Google Cloud’s Gemini for Government products are the first to be offered in the one-stop-shop, and others from xAI, OpenAI and Anthropic are expected to soon follow.
Following the enterprise-wide genAI hub’s release, the military services are pursuing different approaches in their prioritization of the still-maturing capabilities.
The Air Force announced plans shortly after to phase out NIPRGPT, and the Marine Corps is also moving away from legacy products in favor of DOD’s new enterprise offering. The Coast Guard, on the other hand, is hustling to refine and enhance its custom Ask Hamilton platform as it also leans into the GenAI.mil suite.
Like the latter, the Army appears to be maximizing its personnel’s access to a range of options — including but not limited to the specialized CamoGPT research and development platform and the Army Enterprise LLM Workspace for service-wide use — as well as commercial products on the Pentagon-run GenAI.mil workspace that are trained on more non-military, internet-sourced data.
“While CamoGPT was initially an experimental tool that gained traction in the absence of an enterprise AI platform solution, its transition back to a research and development capability upon the release of the Army Enterprise LLM Workspace reflects the Army’s focus on maturing the AI culture across the force,” the Army spokesperson told DefenseScoop. “CamoGPT continues to serve as a vital platform for testing agentic AI and exploring future AI capabilities that can enhance mission effectiveness for the Army and transitioning them to the production environment for use at both the tactical and enterprise levels.”
