OpenAI says actors linked to China-based Moonshot AI spearheaded a campaign to extract its models’ hidden reasoning

What happened

Activity began on July 1st, “initially at a low volume.” OpenAI says operators copied its models’ encrypted reasoning and asked a model in a separate conversation to decrypt it. OpenAI is an artificial intelligence company based in San Francisco.

OpenAI said in a company blog post that people associated with China-based Moonshot AI were at the core of an effort to pull hidden reasoning from its models. The post says that a “coordinated campaign” to extract “protected reasoning” from its models, which was “consistent with adversarial distillation,” occurred in July.

OpenAI isn’t sure whether all of the operators it saw were a single actor. After that came “high-volume spikes” on July 24 and 25, with 16,000 requests using an extraction pattern.