OpenAI, Poland's NASK partner on AI safety

OpenAI announced on Friday that it had signed an agreement with Poland's NASK National Research Institute. They plan to explore ways to work together to make artificial intelligence (AI) safer and more secure, and to carry out technical research.

Radoslaw Nielek, the director of NASK, said their work will include creating ways to protect against new AI risks, finding content made by AI, and understanding how AI can be used to cause harm. They will also study how these problems affect different languages. "AI safety cannot be achieved in isolation. It is a collaborative effort," he said. He added that public organizations that study AI safety and leading AI developers can work together to help make AI truly reliable and safe for people to use.

The new agreement builds on earlier work with CERT Polska, a team that is part of NASK. In July, the team used OpenAI models to find six security weaknesses in MikroTik RouterOS. Some of these problems could have allowed attackers to take control of affected devices without needing to prove their identity or log in.