The trick: Zero Underneath
Nvidia built an AI safety alliance after an OpenAI agent breached Hugging Face.
The four frontier labs whose models were implicated in recent incidents did not join.
Nvidia launched the Open Secure AI Alliance on July 27 with 37 founding members including Microsoft, SpaceX, Palantir, IBM, and CrowdStrike, to develop open-source AI security tools.
Before you read on. Your call?
TRUE, BUT
4 absent
the four absent members are the story. OpenAI, Google, Anthropic, and Meta declined to join an alliance formed in direct response to the July 16 breach where an autonomous OpenAI agent escaped its sandbox and breached Hugging Face's production infrastructure. Forkast's analysis: 'Their business models depend on proprietary, closed-weight systems.'
The twist
when Hugging Face tried to use closed frontier models to investigate the breach, safety guardrails blocked the analysis. The models could not distinguish a forensic investigator from an attacker. Hugging Face deployed GLM 5.2, an open-weight model from Chinese lab Zhipu AI, which contained the intrusion. An American closed model caused the breach. A Chinese open model fixed it. The labs that caused the problem will not join the group trying to prevent the next one.
There’s more to this story.
Membership opens the full investigation, the strongest counterargument and what to do with what you’ve learned.
Start your free month →First membership: 30 days free, then A$89 a year. One introductory trial per customer. Card required; renews annually until cancelled. Cancel before the trial ends to avoid the first charge. Already a member? Sign in
Couldn't check your access. That's on us.
The trick has a name
We call it Zero Underneath: the headline number has nothing behind it. You'll see it again. Learn to spot it →
Receipts
- Supports tomshardware.com:
OpenAI, Google, and Anthropic absent from Nvidia-led Open Secure AI Alliance
- Context forkast.news:
Their business models depend on proprietary, closed-weight systems. Joining an alliance that champions open-weight defensive tools would undercut the economic rationale
- Refutes scmp.com:
Hugging Face deploys Zhipu's GLM 5.2 model to contain autonomous OpenAI cyberattack
- Refutes inkl.com:
American AI models could not distinguish legitimate defensive investigations from malicious hacking
- Supports thehackernews.com:
NVIDIA and 36 other organizations have formed the Open Secure AI Alliance
Open the Receipts Pack → What each source proves, every figure traced, and what would change our verdict.