Hancom said on the 15th that it agreed on a cooperation agenda with Poland's state-certified research and development (R&D) center 7Bulls (7Bulls) and AI/IT corporations Algomine (Algomine) to co-develop an agentic OS.
They will pursue technological cooperation centered on four areas: ▲ productization and localization ▲ integration with existing (legacy) systems ▲ governance and response to European Union (EU) regulations ▲ market cooperation and commercialization.
Hancom plans to develop a European-style sovereign agentic OS by adding AI agents without completely replacing existing systems. It is a structure that connects AI on top of existing work systems to make them intelligent, and, according to the company, public institutions and corporations can adopt AI without large-scale system replacements.
Localization work will start with Polish language support. Hancom is reviewing the use of Poland's own large language model (LLM), "Bielik," and will also jointly develop technologies for Polish agent performance evaluation systems and locale support such as date, currency, and character notation.
The deployment environment will default to closed networks and on-premise. The plan is to implement a structure in which data does not leak externally by enabling AI agents to operate within customers' internal computer networks without going through external clouds.
Integration with existing systems is also a core task. Hancom and 7Bulls will jointly develop connectors that link local mission-critical systems with AI agents. This technology enables AI to read and use data from existing systems. They also plan to push pilot projects targeting platforms such as electronic tax invoices and e-government.
They will also reflect EU regulatory compliance from the development stage. The two companies will prepare development guidelines aligned with the EU AI Act and the General Data Protection Regulation (GDPR), and, using Algomine's local customer network, will pursue joint proofs of concept (PoC) and commercialization targeting Poland's public and financial markets.
This move also coincides with the EU AI Act's implementation schedule. The act's transparency obligations take effect on Aug. 2, and obligations for high-risk AI systems take effect in December 2027. Violations of the rules can result in a penalty surcharge of up to 7% of worldwide revenue.
Hancom CEO Kim Yeon-su said, "Europe's public systems are assets accumulated over decades, and our approach is not to replace them but to make them intelligent," adding, "AI sovereignty is not completed by building infrastructure. You need a layer that makes data actually work on top of it, and Hancom aims to fill that role."