Legion Intelligence to Launch on NIPRGPT, Advancing the U.S. Government from Chat to Agentic Operations

Agents that act—compressing the OODA loop and giving time back to the mission

WASHINGTON, Sept. 22, 2025 (GLOBE NEWSWIRE) -- Legion Intelligence, a leading provider of scalable and secure Generative AI (GenAI) solutions for the Department of War, today announced its agent platform will launch on NIPRGPT this fall, extending the program’s evolution from chat interfaces to agentic systems that execute work under policy. The rollout builds on Legion’s deployments designed for secure, classified, and on‑prem Department of War environments, including enterprise‑level deployments supporting U.S. Special Operations Command.

Unlike basic chat assistants, Legion enables permissioned software agents that call approved tools and data, coordinate with one another, and take auditable actions—shortening the Observe–Orient–Decide–Act (OODA) cycle and reclaiming administrative time.

“Chat has been a great on‑ramp for AI, but agents will be the real force multiplier for the Air Force and the Department of War,” said Ben Van Roo, co-founder and CEO of Legion Intelligence. “Like drones in Ukraine have changed the physical landscape of war, thousands of agents will soon augment workflows across a variety of use cases—from kill chains to staff officers. Our focus is enabling that future, with the software and tooling to build and manage the agentic future.”

Through allow‑listed connectors and data gateways, Legion is designed to embed alongside legacy maintenance, personnel, planning, and operations applications, engage data applications and alerts (e.g., Dataminr), work across productivity (SharePoint, Teams, Jira), and interoperate with systems of record (e.g., Oracle, Palantir)—with row/column controls, provenance, and end‑to‑end audit.

What’s New on NIPRGPT

Agent Workbench & Control Plane — Create, govern, and run agents with workflow design, data integrations, policy enforcement, cross‑enclave scheduling, full observability, and a one‑button kill switch.
Enterprise GenAI Fabric — Embed and share GenAI and agents across applications and workspaces to turn conversation into execution.
Secure Connectors & Data Gateways — Allow‑listed integrations and entitled data paths for REMIS, CAMS, Dataminr, Microsoft SharePoint/Teams, Oracle, and Palantir, with row/column controls and end‑to‑end provenance.
Multi‑Model Routing & Resilience — Task‑aware routing across commercial and open‑source models, plus guardrails, drift alerts, and a continuous red‑team/test range.
Mission Kits (Prebuilt Templates) — Configurable starter agents and policy‑as‑code for intelligence, logistics, and acquisition use cases.

Availability
Legion on NIPRGPT will be available this fall to eligible U.S. Government users on CUI networks. For pilots or production on NIPRNet, classified enclaves, or on‑prem, contact info@legionintel.com.

About Legion Intelligence
Legion Intelligence (formerly Yurts AI) is an agent-orchestration platform for defense, government, and enterprise organizations. Legion empowers teams to securely deploy, manage, and scale intelligent autonomous agents, enabling sophisticated, efficient coordination across sensitive operational environments. By integrating advanced MLOps, robust security protocols, dynamic decision-making, and human-centric design, Legion ensures every critical mission is backed by a capable legion of intelligent agents.

With Legion, the future of secure, sophisticated AI orchestration is here. Existing and prospective customers can confidently navigate their most sensitive environments, leveraging Legion’s proven capabilities to enhance operational effectiveness, security, and strategic decision-making. Visit https://www.legionintel.com/ to learn more.

Media Contact
Carly Bourne
legion@bulleitgroup.com

Legal Disclaimer:

EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.