Insider Brief
- Ondas and Palantir Technologies have formed Regna AI, a standalone company that will develop sovereign AI models for defense and other mission-critical operations.
- Regna AI will use Palantir’s AIP and open-weight models to run mission-specific AI within customer-controlled infrastructure and on edge systems including drones, ground vehicles, surface vessels and command-and-control platforms.
- Ondas, which initially owns and funds Regna AI, will be its first customer, with initial deployments expected over the coming months before expansion into additional defense, security, critical-infrastructure and industrial applications.
Ondas and Palantir Technologies have formed Regna AI, a standalone company that will develop sovereign AI models for defense and other mission-critical operations.
Regna AI is initially wholly owned and funded by Ondas, while Palantir will provide model-training infrastructure and support deployment, according to Ondas. It will also be Regna AI’s first customer, with Ondas using its models for autonomous mission orchestration at the edge.
Ondas said Regna AI will develop fine-tuned AI models designed around specific missions rather than relying solely on general-purpose frontier models. The systems will run within customer-controlled infrastructure and on edge hardware, allowing organizations to keep sensitive data and AI operations inside their own environments.
“No government should have to build its AI strategy on models or data infrastructure it does not control,” Ondas chairman and CEO Eric Brock, Chairman said in the announcement. “Regna AI combines open-weight models, customer-controlled mission data, and Palantir’s operational platform to give defense organizations control over the AI that supports their missions. We believe Regna AI can become a foundational AI platform for the U.S. Department of War, allied governments and the defense and industrial companies that serve them.”
Regna AI initially plans to concentrate on ISR applications before expanding into other defense missions as well as security, critical infrastructure and industrial markets. Initial applications include edge hardware such as drones, stratospheric balloons, ground vehicles and surface vessels, along with command-and-control systems. The models are being designed to operate under denied, degraded, intermittent or limited connectivity.
Ondas indicated Regna AI’s systems will help operators assess information, compare possible actions and propose changes to mission plans, with humans retaining review and authorization.
Palantir will provide the underlying training infrastructure, while the companies plan to use forward-deployed engineers to adapt models to customer workflows and operational requirements. Regna AI will use Palantir’s Artificial Intelligence Platform, or AIP, and open-weight AI models that can be adapted using customer data.
The new company builds on an existing collaboration between Ondas and Palantir around SkyWeaver C4I, an open-architecture system that combines surveillance and edge processing across multiple operating environments. Regna AI plans to use that foundation for mission-specific models deployed across connected and disconnected systems.
“Sovereign AI, trained on mission knowledge and tradecraft, outperforms frontier models while removing dependence on them,” added Palantir co-founder and CEO Alex Karp. “All layers of the stack are owned and controlled by the sovereign institution. Regna AI will put that within reach of institutions that could never justify building a frontier model of their own.”
Ondas said initial customer deployments are expected over the coming months.