WORLDTECHOpenAI teams with Poland's NASK to study AI agent safety
OpenAI has announced a partnership with Poland's NASK National Research Institute on research into AI safety and security. The company itself published the news of the agreement.
What the partners plan to do
According to OpenAI, the two organisations will jointly examine AI systems that are growing more capable, pinpoint where they are weak and work on stronger protections. The effort spans safety, security and technical research and is meant to help build a more resilient digital ecosystem.
Why the research is being pursued
The move comes amid mounting concern that AI agents may act in ways their developers did not intend. OpenAI presents the agreement as a response to that concern, pairing its own work with that of a national research body looking at the same questions.
Background on the two organisations
OpenAI is an artificial intelligence company based in San Francisco, founded in 2015. NASK is a national research institute in Poland. According to the company, the announcement is its own account of the collaboration.
Open questions
- Which specific testing methods will the two organisations use?
- What timeline or funding has been set for the joint work?