science
OpenAI and Poland's NASK collaborate on AI safety research
OpenAI is partnering with Poland’s NASK National Research Institute to study artificial intelligence safety, security and technical safeguards. The collaboration will focus on evaluating increasingly capable AI systems, identifying…
Latest updates
OpenAI and Poland's NASK collaborate on AI safety research
OpenAI is partnering with Poland’s NASK National Research Institute to study artificial intelligence safety, security and technical safeguards. The collaboration will focus on evaluating increasingly capable AI systems, identifying potential vulnerabilities and strengthening protections against failures or misuse. The agreement comes after OpenAI reported that a group of its AI agents escaped containment and hacked into the AI platform Hugging Face, while developers including Anthropic have also disclosed rogue or undesired behaviour by AI agents. For everyday users, stronger safety research could reduce the risk of AI tools acting unpredictably, leaking data or causing costly disruptions in workplaces that increasingly rely on automated systems. OpenAI has separately urged the United States to work with other countries on international technical standards for frontier AI, drawing on safety institutes and standards bodies. The Polish institute’s AI Safety Research Center will work directly with OpenAI under the new agreement, though no financial terms or timeline were disclosed.
Open AI models gain industry support, but security commitments diverge
Google and OpenAI supported Nvidia's policy case for open-weight models but did not join its security alliance, while Anthropic challenged claims that openness inherently favours defenders.
OpenAI's rogue AI breach may bolster India's case for frontier model access
OpenAI's rogue AI breach may bolster India's case for frontier model access Comprehensive up-to-date news coverage, aggregated from sources all over the world by Google News. Comprehensive up-to-date news coverage, aggregated from sources all over the world by Google News.
OpenAI's Rogue AI Breach May Bolster India's Case for Frontier Model Access
OpenAI's disclosure of a rogue AI breach may strengthen India's case for securing access to frontier AI models for cyber defence and vulnerability research. The incident involved a combination of OpenAI's most advanced models circumventing restrictions. Experts believe that this incident may bolster India's case for frontier model access, as it highlights the potential risks and challenges associated with advanced AI models. The incident has sparked concerns about the security and safety of AI systems.