AI Briefing
KOSign in

OpenAI and Poland’s NASK sign agreement to advance AI safety and security research

·2026.10.23 09:00

Key point

The collaboration aims to evaluate capable AI systems and identify vulnerabilities, building on prior work where CERT Polska used OpenAI models to find six MikroTik RouterOS flaws.

Details

  • OpenAI and Poland’s NASK National Research Institute signed an agreement to collaborate on AI safety, security, and technical research.
  • NASK’s AI Safety Research Center, serving as the Polish Artificial Intelligence Safety Institute (AISI), will work with OpenAI to evaluate increasingly capable AI systems and strengthen safeguards.
  • The partnership builds on existing international collaborations with the US Center for Advancing Standards and Innovation for Super Intelligence (CAISSI) and the UK AI Security Institute (AISI).
  • In July, CERT Polska (operating within NASK) used OpenAI’s cyber capable models via Daybreak to discover six vulnerabilities in MikroTik RouterOS, including flaws allowing unauthenticated device control.
  • The collaboration supports OpenAI’s call for international technical standards for frontier AI, leveraging national expertise to assess risks and test safeguards.
openaiai-safetycybersecuritypolandnaskmikrotikcert-polska
Read originalShare on Threads

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.

Related

OpenAI BlogOpenAI publishes blueprint for empowering cyber defenders with advanced AIOpenAI BlogOpenAI Details Unauthorized Model Access to Australian Government Systems and Commits to RemediationTLDR AIJacob Coxon Resigns from Anthropic, Declaring 'Coxon Day' as AI Safety Concerns Go MainstreamHacker NewsOpenAI Discloses Four Rogue Incidents Involving Unauthorized Data Access and Crypto PaymentsRedditGladstone AI Report Warns of Self-Improving AI Race and 3-12 Month RSI TimelineHacker NewsSecurity experts blame frontier AI labs for skipping basic containment in recent agent incidents