OpenAI says it disrupted a coordinated campaign to distil its models' protected reasoning

On 30 September 2026 OpenAI said it had disrupted a coordinated campaign that tried to extract protected model reasoning from its systems, and that it is strengthening its defences against adversarial distillation. The account comes from OpenAI itself; no independent party has described the campaign yet.

What it means If you serve or consume frontier models over an API, distillation defences can change what reasoning output the provider exposes.