Skip to content
NewA calmer, faster ZeroHour

OpenAI hardens GPT-6.2 against distillation of protected reasoning

Original title: Combatting adversarial distillation

OpenAI News ()12 min read1
Part of a story covered by 2 sourcesOpenAI hardens GPT-6.2 against extraction of protected reasoningMerged summary & timeline
AI summary · x-ai/grok-4.7Summaries by Vani AI

In OpenAI's internal red-team, recovering protected reasoning fell from 54.0% on GPT-6 Astra to 0.2% on GPT-6.2 Astra.

OpenAI says unauthorized distillation can copy frontier cyber and biology capabilities into a model that lacks the original safeguards. It reports stronger protections for exact chain-of-thought on GPT-6.2 Astra and GPT-6.2 Sol, combining training, classifiers, API controls, and disruption of coordinated extraction. An internal study found students trained on exact reasoning traces gained more cyber skill than those trained on user-visible summaries.

Unconfirmed: The 54.0% and 0.2% rates are from OpenAI's own internal evaluation, not an independent test.

Key facts
  • Internal adaptive red-team success fell from 54.0% to 0.2% on protected reasoning.
  • Safeguards were strengthened on GPT-6.2 Astra and GPT-6.2 Sol.
  • OpenAI says the controls now align more closely with biology and cybersecurity safeguards.
  • Students were compared after supervised fine-tuning and light reinforcement learning.
Explain this AI news with Vani AIWhat was announced, what is new and why it matters, in a new chat

From the original article

12 min read at openai.com

OpenAI’s layered approach to adversarial distillation combines model training, safeguards, detection, and enforcement to protect against capability extraction.

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.