--- id: "2026-10-23-openai-nask-poland-ai-safety-partnership" url: "https://postcutoff.com/e/2026-10-23-openai-nask-poland-ai-safety-partnership/" as_of: "2026-11-22T13:32:00+01:00" date: "2026-10-23" date_precision: day category: policy-safety importance: 2 confidence: high status: [Confirmed] sources: 2 editor: human editor human_review: null version: "2026-10-29" --- As of: 2026-11-22 13:32 CET. Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/ Canonical page: https://postcutoff.com/e/2026-10-23-openai-nask-poland-ai-safety-partnership/ # OpenAI partners with Poland's NASK, whose AI Safety Research Center acts as the Polish AI Safety Institute, on AI safety and security research OpenAI announced on Oct 23, 2026 a partnership with NASK, Poland's National Research Institute for cybersecurity and AI safety, for exploratory research on AI safety and security: ways to evaluate increasingly capable systems, finding security vulnerabilities and strengthening safeguards. NASK's AI Safety Research Center serves as Poland's AI Safety Institute. It builds on CERT Polska, NASK's emergency-response team, using OpenAI models in July to find six vulnerabilities in MikroTik RouterOS, since fixed. ## Key facts - Announced: OpenAI post 'OpenAI and Poland's NASK partner to advance AI safety and security' (Oct 23; openai.com returns 403 to our fetcher); confirmed on LinkedIn by George Osborne, head of OpenAI for Countries (EdTech Innovation Hub, Oct 29) - Scope (EdTech Innovation Hub): exploratory research on evaluating increasingly capable AI systems, identifying security vulnerabilities and strengthening safeguards; NASK also works on multilingual security challenges - Prior work: in July 2026 CERT Polska used OpenAI models to find six vulnerabilities in MikroTik RouterOS; MikroTik has released fixes - Quotes: NASK director Radosław Nielek: 'AI safety cannot be achieved in isolation'; OpenAI's Tom Duff Gordon (VP, head of EMEA policy): 'As AI becomes more capable, it's vital that we draw on expertise beyond our own organization' ## What happened A frontier lab signed a research agreement with a national AI safety institute of an EU member state that is not one of the large AI powers, building on a concrete earlier result: vulnerabilities found with OpenAI's models in widely used router software. ## Why it matters It adds Poland's institute to the national AI safety institutes that work directly with a frontier lab on evaluations and cybersecurity, with a track record of real vulnerabilities found and fixed. ## Your AI and this story - GPT-6.2 Astra (training cutoff May 2026): 145 days after its cutoff - Claude Opus 5.5 (training cutoff June 2026): 115 days after its cutoff - Gemini 4 Hydrogen (training cutoff March 2026): 206 days after its cutoff - Grok 4.7 (training cutoff May 2026): 145 days after its cutoff ## Sources 1. [OpenAI: OpenAI and Poland's NASK partner to advance AI safety and security (Oct 23; 403 for us)](https://openai.com/index/openai-nask-ai-safety-security-partnership) (openai.com, official) 2. [EdTech Innovation Hub: OpenAI signs AI safety agreement with Poland's NASK research institute (Oct 29)](https://www.edtechinnovationhub.com/news/openai-signs-ai-safety-agreement-with-polands-nask-research-institute) (edtechinnovationhub.com, press) ## Changes - 2026-10-29 (filed): Created (OpenAI RSS entry; EdTech Innovation Hub read directly; skipped twice as minor before) (news agent) ## Related - 2026-10-21: [PHP2Shell: critical remote-code-execution flaws in PHP, found with GPT-5.6 Sol and Gemini 4 Argon, force emergency patching and take websites offline worldwide](https://postcutoff.com/e/2026-10-21-php2shell-ai-found-php-rce/index.md)