
What happened
OpenAI said internal adaptive red-teaming showed the average attack success rate for recovering protected reasoning fell from 54.0% against GPT-6 Astra with its original safeguards to 0.2% against GPT-6.2 Astra with updated safeguards.
Why it matters
OpenAI frames adversarial distillation (using a frontier model's outputs to train another model without its safeguards) as a safety and national security risk, so stronger protection against reasoning-trace extraction may make it harder to copy advanced capabilities, though OpenAI plans to share more about recent campaigns.
WHO IT HITSModel developers and platform operators offering frontier AI models, plus their security and policy teams, may need to adopt similar request-level and campaign-level defenses against unauthorized capability extraction.
Summaries like this, every morning.
OpenAI's disclosure links technical safeguards to a broader policy argument. The company says adversarial distillation lets a second model learn advanced frontier capabilities without inheriting the safeguards that govern how the original model can use them, and it describes the concern as especially clear in dual-use domains like cybersecurity and biology. Its approach combines adversarial testing, model training, classifiers, and API-level protections with investigations into coordinated extraction attempts.
The company says it has strengthened protections for protected reasoning in GPT-6.2 Astra and GPT-6.2 Sol, tying a model's protected reasoning to the conversation history that produced it and blocking older model families with weaker safeguards from reading it. It also says an internal adaptive evaluation using strong automated red-teaming found the average attack success rate for recovering protected reasoning fell from 54.0% to 0.2%.
OpenAI connects the technical work to national security, citing a February briefing to Congress on activity indicating that DeepSeek was attempting to distill U.S. frontier models and a September report attributing a core part of a disrupted extraction campaign to individuals associated with Moonshot AI, the developer of Kimi. It says it supports legislation and export controls, and it recommends steps other developers can take, such as keeping successful extraction attack logs as regression tests and combining request-level mitigations with campaign-level enforcement.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Apple is hiring a senior engineer to lead its own device-controlling AI agent, competing with Muse, Dots, Duet…

Nvidia is nearing a deal to acquire AI startup Inception for at least $500 million, per three people told The…

Underwriters bought the full 147 million-share overallotment at $83, adding $12.2 billion and pushing Anthropi…

Oxide Computer Co., which raised $445 million in October to expand rack production, built its business on the…
OpenAI now requires gateways and routers that provide third-party access to its AI models to be reviewed and a…

Anthropic shares traded as low as $83.15 on Friday, down more than 14% on their second day of trading, leaving…
