The Open Secure AI Alliance comes in the wake of a cybersecurity incident in which OpenAI models targeted tech company Hugging Face, according to a Monday (July 27) Nvidia blog post.
“That incident showed a practical truth: when defenders cannot inspect, adapt and run advanced AI on their own infrastructure, their ability to respond is constrained at exactly the moment speed matters most,” the post said. “Companies and countries need open frontier defensive tools and techniques so critical industries can build security systems across a multi-vendor ecosystem and avoid single points of failure.”
The alliance includes Capital One, CrowdStrike, DoorDash, Microsoft, IBM, SpaceX and others, according to the post. It will strive to identify and remediate vulnerabilities using open technologies.
“Just as open source created a shared foundation for software, the United States and its partners now face a choice in AI security: whether the defenses that protect our infrastructure will sit inside a few opaque systems or be built on open models, harnesses and tools that any defender can study, adapt and deploy,” the post said.
The world needs closed and open AI models, according to the post. The latter is essential for cybersecurity, as they “democratize defensive capabilities, increase transparency for defenders, enable cyber defense while protecting data, and complement frontier closed models with customizable, localized controls.”
But like any technology, open models can be misused, including via efforts to hinder safeguards or reconfigure capabilities to conduct cyberattacks, the post said.
“Across the Alliance, contributors are building an open defense stack for agents,” including Microsoft’s MDASH and SpaceXAI’s open sourcing of the Grok Build coding agent, according to the post.
The group is calling on companies and governments to “invest in shared open infrastructure for AI defense—datasets, evaluation frameworks, attack simulators and red-teaming tools—much as past generations invested in open source software,” the post said.
OpenAI announced last week that a cybersecurity incident reported by Hugging Face earlier in the month was caused by OpenAI’s models during cyber capabilities testing.
“We consider this incident to be an unprecedented cyber incident, involving state-of-the-art cyber capabilities, and are responding accordingly,” OpenAI said Tuesday (July 21).
For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.
