Nvidia has launched the Open Secure AI Alliance, a new global initiative bringing together around 40 leading technology companies to strengthen the security of artificial intelligence systems and address emerging vulnerabilities in open-source and autonomous AI models.
The alliance includes major industry players such as Microsoft, Dell Technologies, CrowdStrike, SpaceX, and Hugging Face, reflecting a collaborative effort to establish stronger security standards as AI adoption accelerates worldwide.
Alongside the announcement, Nvidia introduced a new open-source software tool on GitHub designed to help developers monitor the behaviour of AI agents, detect abnormal activity, and reduce the risk of unintended or malicious commands.
The initiative comes amid growing concerns over AI security following a previous incident involving OpenAI models that affected the Hugging Face platform. The event highlighted the need for more robust safeguards to protect increasingly advanced AI systems from evolving cyber threats.
Through the Open Secure AI Alliance, Nvidia and its partners aim to promote collaboration, improve AI security practices, and support the development of safer and more reliable AI technologies for developers, businesses, and users.
