EIMIN: €12 million for AI solutions for the Lithuanian language
In order to strengthen Lithuania's national security and technological progress, as well as to maintain the accuracy of the Lithuanian language, the Ministry of the Economy and Innovation is funding the development of artificial intelligence solutions for the Lithuanian language. More than €12 million has been earmarked for the implementation of 6 projects.
"Lithuanian language resources are essential for a wide range of national security applications. They can facilitate surveillance, intelligence analysis and cyber security. By investing in Lithuanian language resources for artificial intelligence, we will not only strengthen our capabilities in these areas, but also better protect our citizens from emerging threats," said Aušrinė Armonaitė, the Minister of the Economy and Innovation.
6 calls have been launched to develop solutions for AI technologies adapted to the Lithuanian language. The funds will be allocated to the Information Society Development Committee, which will carry out public tenders for the projects. These tenders are expected to be launched in May-June and will be open to companies.
"Bringing AI technologies adapted to the Lithuanian language closer to the needs of business, science and society will ensure their accessibility and make it easier and simpler to communicate and use various services. The resources to be developed will range from identifying false information and facilitating the use of the Ukrainian language to accelerating the development of medicine, biotechnology and genetics," says Erika Kurochkina, Deputy Minister of the Economy and Innovation.
The first project will develop a common Lithuanian language text database and vectorised Lithuanian language models. This will involve an investment of €4.8 million. The lexicon is one of the fundamental resources of Lithuanian language technology, as its completeness, quality and lexical diversity determine the quality and usability of the intellectual technology solutions developed. A vectorised model of the Lithuanian language would also enable the design and development of AI-based solutions and innovations in areas related to human language.
The second project will develop a text database for automatic identification of false information. This will involve an investment of €3.7 million. This text database will enable the design and development of AI-based solutions to identify the content of false information in online media. This will contribute to the protection of democratic values and the strengthening of civil society.
The third project concerns the updating of monolingual and multilingual text databases. It has received €1.1 million in EU funding. This project will also collect and process parallel and monolingual texts in English, French, German, Russian and Polish. These will be made publicly available and can be used to develop new machine translation systems or other services based on machine translation and artificial intelligence. This will be of particular benefit to public authorities and businesses, who will be able to use machine translation free of charge, integrate these text databases into their websites and facilitate the translation of documents.
The fourth project, with a budget of €1.1 million, will develop multilingual and monolingual text databases for 5 more languages: Spanish, Ukrainian, Norwegian, Swedish and Danish. There is currently a strong demand in the market for solutions tailored to the Ukrainian language. Ukrainians living in Lithuania want to understand information written in Lithuanian and receive services in their own language, and businesses are increasingly cooperating with and investing in Ukrainian companies.
The fifth project will develop synthetic parallel text databases. It has been allocated €660 000. These text databases will be used to adapt machine translation to general or medical domains, such as medical language translation systems, translation plug-ins for websites, plug-ins for professional translators, etc.
The sixth project, "Development of a Human Phenotype Ontology in Lithuanian", will codify terms and concepts describing the diversity of human phenotypes (observable changes in characteristics or sets of characteristics of an organism). A budget of €740 000 has been allocated. The project focuses on solutions for the field of genetics. It will improve interdisciplinary communication and increase international communication in genetics.
All project activities must be completed by 30 April 2026.
The full terms and conditions of the call are available on the European Union Investment website.
The measure is implemented within the framework of the Economic Recovery and Resilience Plan "Next Generation Lithuania", funded by the European Union's NextGenerationEU tool for economic recovery and resilience.
