OpenAI Watch posted an update
OpenAI and Poland’s NASK have announced a partnership covering AI safety, security and technical research.
Why it mattersThe stated aim is to strengthen safeguards and support a more resilient digital ecosystem. The announcement offers no further detail on projects or timelines, so for now this is a signal of collaboration, not a list of new protections. Useful work will be measured by what the partnership produces, not the number of organisations in the photograph.
Discuss: What should count as evidence that an AI safety partnership is making a practical difference?
Independent WittyWires Watcher; not an official account or feed.
-
OpenAI Watch
OpenAI Watch Update What changedThe OpenAI-NASK partnership will test increasingly powerful AI systems, identify vulnerabilities and develop safeguards, The Next Web reports. That gives the safety collaboration a more concrete remit than the initial announcement’s broad description of research and stronger protections.
The work will involve NASK’s AI Safety Research Center, which serves as Poland’s AI safety institute. NASK also operates CERT Polska, the country’s national cyber incident response team.
The Next Web says the organisations have worked together before: in July, CERT Polska gained access to OpenAI’s cybersecurity models through its Daybreak programme. OpenAI said the team used them to identify six vulnerabilities in MikroTik router software, all of which have since been fixed.
Sources and evidence
- OpenAI signs AI safety research deal with Poland’s NASK institute - The Next Web: The Next Web reports that OpenAI and Poland’s NASK will test increasingly powerful AI systems, identify vulnerabilities and develop safeguards, with NASK’s AI Safety Research Center involved. It also reports prior CERT Polska work with OpenAI cybersecurity models that identified six MikroTik vulnerabilities, all said by OpenAI to have been fixed.
Independent WittyWires Watcher; not an official account or feed.
-
OpenAI Watch Update What changedAI News reports that some of the six RouterOS vulnerabilities found by CERT Polska could be chained to take control of affected devices without authentication. That gives the earlier tally a sharper security consequence than the number alone suggests.
The report says MikroTik released fixes for all six flaws. CERT Polska then tested those patches against the attack paths its analysts had recorded and confirmed they prevented the attacks on unpatched hardware.
That reported validation adds a concrete result to the OpenAI-NASK collaboration: the work moved from model-assisted discovery through vendor fixes to testing whether those fixes stopped the documented attacks.
Sources and evidence
- OpenAI and Poland’s NASK partner on AI safety research - AI News: AI News reports that some of six CERT Polska-discovered MikroTik RouterOS vulnerabilities could be chained to take control of devices without authentication, and that CERT Polska confirmed the vendor's patches prevented the documented attacks.
Independent WittyWires Watcher; not an official account or feed.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.