Ondas unveils Regna AI, saying it is powered by Palantir technology to target governments and “sovereign AI” use cases

Ondas says it has launched Regna AI, a new artificial intelligence offering intended for governments and enterprises that want more control over how AI is built and deployed. In its announcement, the company described Regna AI as bringing “mission-specialized” fine-tuned AI models to customers, and it said the system is “powered by Palantir.”

The release was published on October 12, 2026, alongside the company’s listing as Ondas, Inc., on the Nasdaq under the ticker ONDS. The same post frames the initiative as part of a broader effort to advance “sovereign AI,” a term commonly used in the market to refer to AI systems that are developed and operated with an emphasis on jurisdictional control, security, and ownership of data and deployment environments.

In practical terms, Ondas is positioning Regna AI around customization. The company’s language emphasizes that its models are fine-tuned for specific missions, rather than being one-size-fits-all chat or analytics tools. Fine-tuning generally means adjusting a model’s behavior for a narrower set of tasks, which is often a prerequisite for use cases like operational decision support, risk screening, or domain-specific assistance.

Ondas did not provide, in the announcement text available here, additional technical details such as the specific model types used, the training approach, data handling methods, deployment options, or whether Regna AI will run in particular environments (for example, on-premises, in private cloud, or through a managed platform). It also did not disclose contract terms, customer names, or pilot timelines.

The company’s choice to cite Palantir reflects Palantir’s broader market presence in data integration and applied AI deployments for regulated or mission-driven organizations. Palantir is widely known in the sector for working with customers that need traceability, governance, and operational integration, but Ondas’ announcement, as provided, does not specify which Palantir products or modules are involved.

Sovereign AI has become a major theme for both governments and large enterprises as AI adoption accelerates and regulators focus on data protection, auditability, and risk management. Buyers looking for sovereignty often weigh whether they can keep sensitive datasets within approved jurisdictions, maintain visibility into system behavior, and control vendors or deployment paths.

Still, what remains unclear from the announcement is how Regna AI will be evaluated and governed once deployed. For example, the release did not outline performance metrics, compliance frameworks, audit procedures, or how updates to fine-tuned models would be managed over time.

Investors and customers may look next for additional details on implementation, such as references to initial customers or pilots, information on how the fine-tuned models are produced and refreshed, and any clarity on governance and security controls that support “sovereign” requirements. The absence of those specifics means the commercial and technical scope of Regna AI remains to be confirmed as the program develops.