OpenAI and Poland's NASK Partner on AI Safety Research

OpenAI and NASK, Poland’s national research institute, have entered a partnership focused on AI safety research. The agreement establishes a framework for joint work on evaluating and testing artificial intelligence models.

The collaboration responds to growing demand from governments for domestic capacity to assess AI systems. It also addresses the need for evidence that global models operate safely within specific national contexts. The work will center on research that helps determine how AI behaves under conditions relevant to Polish institutions and public services.

NASK brings its role as a state research body to the partnership, while OpenAI contributes its AI development expertise. The arrangement signals an effort to build local technical capability for safety evaluation, rather than relying solely on external assessments.