Nvidia has teamed up with 44 tech heavyweights including Microsoft, IBM, Dell, Palantir and Hugging Face, to launch the Open Secure AI Alliance.
The coalition aims to build and share open-source security tools and open-weight models to protect global infrastructure from software vulnerabilities and rogue AI threats.
Incident triggers urgent action
The initiative follows a recent cybersecurity event at Hugging Face, where proprietary commercial tools failed to help during a live intrusion. Defensive teams had to rely on an open-weight model running on their own servers to analyze thousands of suspicious actions and stop the breach.
Reflecting on the event, Nvidia explained, “When closed AI tools – unable to distinguish attackers from defenders – blocked essential forensic analysis, Hugging Face ran the open-weight GLM 5.2 model on its own infrastructure to analyze more than 17,000 actions and contain the intrusion.”
Member companies share essential security tools
To strengthen cybersecurity defenses, member companies are donating software tools directly to the alliance.
Nvidia is contributing its open agent testing framework, Microsoft is offering an automated bug-scanning system and Hugging Face is sharing its safe model storage format.
Pushing back against heavy regulations
The alliance is also calling on global governments to view open AI systems as defensive assets rather than security liabilities. Emphasizing the need for balance, Nvidia stated, “The right response is not to deny defenders access to capable open systems. It is to pair openness with strong safeguards, clear rules against malicious misuse, rigorous evaluation and rapid remediation.”
