US AI firms collaborate to counter unauthorized model distillation by Chinese companies
OpenAI, Anthropic, Google, and Microsoft are collaborating through the Frontier Model Forum to combat unauthorized distillation of their AI models by Chinese companies. US officials reportedly estimate that such unauthorized distillation costs Silicon Valley labs billions of dollars annually, with concerns focused on firms like DeepSeek allegedly extracting model capabilities without permission.
What happened
OpenAI, Anthropic, Google, and Microsoft are sharing information through the Frontier Model Forum to detect and counter unauthorized distillation of US AI models by Chinese companies.
US officials have reportedly estimated that unauthorized distillation costs Silicon Valley labs billions of dollars in lost annual profit. The issue gained attention in January 2025 when Chinese startup DeepSeek released its R1 reasoning model.
Reported impact
OpenAI told US lawmakers that DeepSeek had continued using increasingly sophisticated tactics to pull results from its models despite tightened safeguards and accused the Chinese firm of trying to 'free-ride on the capabilities developed by OpenAI and other US frontier labs.'
Index analysis
Distillation is a legitimate technique where an older 'teacher' AI model trains a newer, more efficient 'student' model at a fraction of the original development cost. The problem arises when third parties, particularly those in adversary nations, use the technique without authorization to replicate proprietary work.
Governance controls
OpenAI, Anthropic and Google have set aside competitive differences to tackle the shared problem of unauthorized distillation. The companies are collaborating through the Frontier Model Forum, an industry nonprofit co-founded with Microsoft in 2023.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.