OpenAI and Poland's NASK Join Forces to Advance AI Safety and Security
OpenAI has signed a formal agreement with Poland's NASK (National Research Institute) to explore collaboration on AI safety, security, and technical research. The partnership pools the expertise of both organizations to better understand emerging risks, strengthen protective measures, and investigate how advanced AI can be leveraged to build a safer and more resilient digital ecosystem.
As the capabilities of AI systems continue to grow, international cooperation between developers and public research institutions is becoming essential for understanding the risks posed by these systems and establishing more robust safeguards. OpenAI's existing collaborations—such as those with the Center for AI Safety and Security Innovation (CAISSI) in the U.S., the AI Safety Institute (AISI) in the U.K., and other public research institutions globally—have already helped strengthen defenses, identify and remediate security vulnerabilities in AI agents, and advance the field of frontier AI safety.
NASK AI Safety Research Center to Collaborate with OpenAI
Under the agreement, the NASK AI Safety Research Center—which serves the functions of Poland's AI Safety Institute (AISI)—will work with OpenAI on AI safety research. This includes exploring methods to evaluate increasingly capable AI systems, identify potential vulnerabilities, and bolster protective measures.
OpenAI has previously advocated for the U.S. to collaborate with nations worldwide—leveraging the expertise of AI safety institutes and standards bodies—to develop international technical standards for frontier AI. Adopting common methodologies for capability evaluation, risk assessment, safety testing, and incident reporting will enable countries to compare evidence and address cross-border risks. The partnership with NASK underscores the value of integrating national expertise into this broader international effort.
Why National-Level Partners Like NASK Matter
NASK's specialized capabilities stem from extensive research into AI safety, which is applied directly to protecting public services and Poland's digital information ecosystem. Without national-level partners like NASK, global AI safety standards might fail to align with the real-world environments in which AI is deployed.
This agreement also builds upon existing cooperation between OpenAI and Polish cybersecurity professionals. In July, CERT Polska—a unit within NASK—gained access to OpenAI's models with cybersecurity-specific capabilities through the "Daybreak" initiative. The team leveraged OpenAI models to discover six vulnerabilities in MikroTik RouterOS. By chaining some of these vulnerabilities together, an attacker could gain control of an affected device without authentication. The vendor subsequently released patches, and CERT Polska has confirmed that these fixes effectively block the attacks they observed.
Accessing OpenAI Tools for Research and Development
As more organizations explore OpenAI's models for safety research, development, and cybersecurity work, reliable access to accounts remains important. Many users search for ways to Buy OpenAI GPT-3 Access or find a Cheap ChatGPT Plus Account to test capabilities without incurring high costs. Platforms like z2u.com offer a marketplace where users can explore GPT-3 Accounts and other OpenAI-related options. Whether you are a researcher evaluating AI safety methods, a developer building new tools, or a cybersecurity professional testing model behavior, services on z2u.com can provide a convenient way to access the accounts you need.
A Shared Commitment to Safer AI
The OpenAI–NASK partnership reflects a growing recognition that AI safety is a global challenge requiring coordinated action. By combining national expertise with frontier AI research, both organizations aim to strengthen safeguards, improve risk assessment, and support the development of international technical standards. As AI continues to evolve, collaborations like this will play a critical role in building a safer and more resilient digital future.