DeepSeek and Moonshot reportedly train models on GPT-6 Astra outputs
Reports say DeepSeek and Moonshot AI use GPT-6 Astra outputs to train new models after Anthropic tightened defenses against distillation.
DeepSeek and Moonshot AI are reportedly training their own new models on outputs from OpenAI's GPT-6 Astra. The practice is known as distillation, and Anthropic had tightened its defenses against it.
OpenAI said it found a coordinated Moonshot-linked campaign in September.
What changed
DeepSeek and Moonshot AI are reportedly training new models on outputs from GPT-6 Astra through distillation.
Type: Security · Applies to: GPT-6 Astra
Why it matters
Model-output theft is emerging as a competitive and security issue, which could lead to stricter access controls and terms of service across frontier model providers.
What you can do
Review API terms and output-use rules if your products depend on frontier model outputs.
Original source
We have no link to the original source for this story yet.
Where we saw it
Forward Future (Matthew Berman)
AI wrote this summary from the sources listed, and nobody reviewed it individually before publication. Check the original source before you rely on it. How it works