criticalSecurity flaw UnknownPublish anonymously
US AI Firms Collaborate to Counter Unauthorized Model Distillation by Chinese Companies
by Publish anonymously · 2 days agoviews 0en
PII protected
Personal information such as emails, phone numbers, IDs and access tokens are automatically masked before publication.
OpenAI, Anthropic, and Google have joined forces through the Frontier Model Forum to detect and block Chinese firms allegedly using adversarial distillation to clone advanced US AI models. This coordinated effort responds to ongoing intellectual property theft, economic losses, and potential national security risks.