Ukraine developing independent AI system with Google open technology, ministry says

Ukraine is developing a sovereign large language model based on Google’s open-weight Gemma framework to achieve strategic independence in artificial intelligence. This initiative aims to serve the public and private sectors, addressing specific linguistic nuances and security needs that foreign proprietary systems often fail to handle. By training the model initially on Google’s infrastructure before transitioning to local servers, Ukraine ensures full control over the technology, reducing reliance on expensive or potentially unsafe foreign services. The project is directly relevant to open data and open-source technologies because it leverages open-weight models to build a custom solution that respects national sovereignty. The development process involves collecting data from over ninety government institutions, including court registries and archives, to ensure the model accurately represents Ukrainian history, law, and diverse linguistic contexts. This approach demonstrates how open frameworks can be adapted for critical national infrastructure, allowing smaller nations to create specialized AI tools that are culturally and legally aligned with their specific requirements, rather than relying on generic global models. Ultimately, this effort highlights the growing importance of open-source AI in maintaining national security and digital autonomy. With military applications already in use for reconnaissance and the new model intended for battlefield management, Ukraine seeks to mitigate risks associated with proprietary systems and cyber threats. The project serves as a case study for how open technologies can empower nations to develop resilient, independent AI ecosystems that prioritize local data integrity and security, offering a viable alternative to dominant foreign tech giants.

Source: thestar.com.my
Published on 2025-12-02

Ukraine developing independent AI system with Google open technology, ministry says

Related news