The initiative supports Veon and Kyivstar’s broader pledge to invest $1 billion into rebuilding Ukraine’s digital infrastructure from 2023 to 2027.
The LLM project aims to strengthen Ukraine’s technological sovereignty by creating a foundation for AI that reflects the country’s language, culture, and data privacy needs.
Developed using open-source architectures and trained entirely on Ukrainian data, the LLM will capture the full spectrum of dialects, historical context, and national terminology.
Crucially, all data will be securely processed and stored within Ukraine, an essential requirement for critical sectors such as healthcare, defence, government, and finance.
Beyond its cultural and linguistic relevance, the model is expected to underpin a range of AI-driven services, from legal and regulatory analysis to tailored applications for education, finance, and healthcare.
A first release is scheduled for December 2025, in collaboration with Ukraine’s Ministry of Digital Transformation and the WINWIN AI Centre of Excellence.
“Today’s agreement is major milestone in building a sovereign, secure and culturally relevant AI ecosystem to drive economic growth, while empowering Ukrainians with tools that enhance their lives,” said Kaan Terzioglu, CEO of Veon Group.
“Ukrainian LLM will empower users to access augmented intelligence tools with the full cultural context and depth of their native language and national resources.”
Oleksandr Komarov, CEO of Kyivstar, added: “Leading the development of Ukraine’s national LLM is aligned with our vision as VEON’s digital operator in Ukraine. With Kyivstar’s deep technological expertise, AI capabilities, and commitment to national development, we are proud to contribute to Ukraine’s digital sovereignty.”
The project follows VEON’s other language-focused AI initiatives, including KazLLM in Kazakhstan and an Urdu-language LLM in Pakistan.
RELATED STORIES
