OpenAI and Poland's NASK sign AI safety and security agreement
OpenAI and Poland's NASK National Research Institute have signed an agreement to explore collaboration on AI safety, security and technical research. According to OpenAI, the partnership aims to better understand emerging risks and strengthen safeguards.
- OpenAI and Poland's NASK National Research Institute signed an agreement to explore collaboration on AI safety, security and technical research.
- NASK's AI Safety Research Center, serving as the Polish AI Safety Institute, will work with OpenAI on evaluating capable AI systems and strengthening safeguards.
- OpenAI says it already works with the US CAISSI and the UK AI Security Institute, and calls for international technical standards for frontier AI.
- In July, CERT Polska used OpenAI models via Daybreak to help discover six vulnerabilities in MikroTik RouterOS, and the vendor released fixes.
What the agreement covers
Under the agreement, NASK's AI Safety Research Center, which serves as the Polish Artificial Intelligence Safety Institute (AISI), will work with OpenAI on AI safety research. The two sides plan to explore ways to evaluate increasingly capable AI systems, identify potential vulnerabilities and strengthen safeguards.
OpenAI says the partnership will bring together expertise from both organizations. Beyond risk and safeguards, it will also look at how advanced AI can contribute to a more secure and resilient digital ecosystem.
Part of a wider push for international standards
OpenAI frames the deal as part of a broader effort. The company argues that as AI systems become more capable, international collaboration between developers and public research institutions will be essential to understanding their risks and building stronger safeguards.
It points to its existing work with the US Center for Advancing Standards and Innovation for Super Intelligence (CAISSI), the UK AI Security Institute (AISI) and other public research institutions around the world. According to OpenAI, that work has already helped strengthen safeguards, identify and fix security vulnerabilities in AI agents, and advance frontier AI safety.
OpenAI has also called for the US to work with countries around the world on international technical standards for frontier AI, drawing on the expertise of AI safety institutes and standards bodies. In its view, common approaches to evaluating capabilities, assessing risks, testing safeguards and reporting incidents can help countries compare evidence and respond to risks that cross borders.
The company says the NASK collaboration reflects the value of bringing national expertise into this effort. It describes NASK's expertise as resting on extensive AI safety research, applied directly to protecting public services and Poland's digital information ecosystem. OpenAI adds that without national partners like NASK, global AI safety standards may remain detached from the realities in which AI is actually used.
Building on CERT Polska's vulnerability work
The agreement also builds on OpenAI's existing work with cyber defenders in Poland. In July, CERT Polska, which operates within NASK, gained access to OpenAI's cyber capable models through Daybreak.
According to OpenAI, the team used those models to help discover six vulnerabilities in MikroTik RouterOS. Some of the flaws could be combined to take control of affected devices without authentication. The vendor released fixes, which CERT Polska confirmed prevented the attacks it observed.
What it means
The agreement is described as an exploration of collaboration, and the announcement gives no timeline, funding or specific projects. What it does indicate is OpenAI's intent to work with national AI safety institutes outside the US and UK, with Poland now added to that group. For NASK, the partnership offers a channel to work with a leading AI developer on evaluation methods and safeguards, building on the security results already achieved by CERT Polska.

