OpenAI, Poland's NASK Sign AI Safety Pact, Building on 6 RouterOS Vulnerability Finds
Updated
Updated · OpenAI · Oct 23
OpenAI, Poland's NASK Sign AI Safety Pact, Building on 6 RouterOS Vulnerability Finds
3 articles · Updated · OpenAI · Oct 23
Summary
OpenAI and Poland’s NASK signed an agreement to collaborate on AI safety, security and technical research, centering on how to evaluate more capable AI systems, identify vulnerabilities and strengthen safeguards.
NASK’s AI Safety Research Center—serving as Poland’s AI Safety Institute—will work with OpenAI on risk assessment and testing aimed at protecting public services and the country’s digital information ecosystem.
The deal extends OpenAI’s broader push for international frontier-AI standards and follows its work with public institutes including the US CAISSI and UK AISI on safeguards and security flaws.
In July, CERT Polska, which operates within NASK, used OpenAI cyber models through Daybreak to uncover 6 MikroTik RouterOS vulnerabilities; the vendor patched them and CERT confirmed the observed attacks were blocked.