# OpenAI says GPT-6.2 Astra leaks protected reasoning in 0.2% of adaptive extraction attacks, down from 54% on GPT-6 Astra

The company says it now acts against thousands to tens of thousands of accounts a week and shares indicators of compromise with other model providers.

- Published: 2026-11-21T06:18:37.672Z
- Canonical: https://polylog.news/ai/2026-11-21/openai-says-gpt-6-2-astra-leaks-protected-reasoning-in-0-2-o
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [OpenAI News](https://openai.com/index/combatting-adversarial-distillation/), [OpenAI (Traditional Chinese)](https://openai.com/zh-Hant/index/disrupting-a-coordinated-model-distillation-campaign/), [Bloomberg](https://www.bloomberg.com/news/articles/2026-02-12/openai-accuses-deepseek-of-distilling-us-models-to-gain-an-edge)

OpenAI published its defenses against adversarial distillation on November 20. Adversarial distillation means extracting a model's capabilities, usually its hidden reasoning, to train a competing model. The post describes defenses at two le…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-11-21/openai-says-gpt-6-2-astra-leaks-protected-reasoning-in-0-2-o (subscription information: https://polylog.news/pricing).