US tech giants OpenAI, Anthropic, and Google have joined forces to combat the growing threat of AI model copying in China. The unprecedented collaboration aims to detect and prevent the unauthorized use of their cutting-edge AI models by Chinese competitors.
The firms are sharing information through the Frontier Model Forum, an industry non-profit, to identify and counter 'adversarial distillation' attempts that violate their terms of service.
US officials estimate that unauthorized distillation costs Silicon Valley labs billions of dollars in annual profit, with some Chinese firms using the technique to replicate proprietary work without authorization.
The move echoes a standard practice in the cybersecurity industry, where firms regularly swap data on attacks and adversaries' tactics to strengthen network defenses.
While the information-sharing effort is still in its early stages, it marks a significant step towards addressing the growing concern of AI model copying in China.
