In a nutshell: Open AI models are essential for cybersecurity because defenders can transparently examine systems and run them on their own infrastructure — an advantage that closed systems do not offer.
Leading companies from cloud, cybersecurity and open-source communities are founding the Open Secure AI Alliance to provide open AI models and tools for cyberdefense. The initiative argues that open, verifiable systems are essential for defenders — both to create transparency and to avoid dependence on a few proprietary solutions.
The Open Secure AI Alliance builds on the work of the Linux Foundation and the OpenSSF and unites over 40 partner organizations, including NVIDIA, Adobe, Cisco, CrowdStrike, Elastic, Hugging Face, IBM, Microsoft, Palo Alto Networks and Salesforce. The goal is to develop open technologies, techniques and tools to protect software and AI agents in the age of artificial intelligence.
The core argument: While closed, proprietary AI systems are valuable for certain tasks, open models are essential for cybersecurity. They democratize defensive capabilities, enable transparency, allow data gatekeeping and provide customizable, localized controls. Open models avoid single points of failure that arise from dependence on a few vendors — a critical risk for critical infrastructure.
A concrete example from the Hugging Face security incident illustrates this: When closed AI tools blocked forensic analysis and could not distinguish between attackers and defenders, Hugging Face deployed the open GLM 5.2 model on its own infrastructure to analyze over 17,000 actions and isolate the intruder. Defenders would not have achieved this critical response speed without access to inspectable, locally runnable AI systems.
The Alliance acknowledges that open models, like any powerful technology, can be misused — by weakening protective measures or repurposing for cyberattacks. However, these risks are not limited to open systems and must be managed wherever advanced AI is deployed. The answer is not to deny defenders access to capable open systems, but to pair openness with strong protective measures, clear rules against malicious use, rigorous evaluation and rapid bug fixing.
Source: blogs.nvidia.com · Published July 27, 2026
Lumi AI News — AI-assisted curation in accordance with Art. 50 EU AI Act. Paraphrasing and classification by Lumi News Pipeline v1.7.3.