
What happened
OpenAI says attacks pulling protected reasoning from prior requests fell from a 54.0 percent success rate on GPT-6 Astra to 0.2 percent on GPT-6.2 Astra; active-request attacks dropped from 22.9 to 0.9 percent.
Why it matters
The company is also extending selected improvements to GPT-6 Astra, so the protection is not confined to the newest models.
WHO IT HITSOperators of AI models and the security teams defending them are the most directly affected, since the same attack techniques that aim to lift hidden reasoning could be used to copy proprietary model behavior.
Summaries like this, every morning.
OpenAI's new figures describe a sharp drop in the success of attacks that try to recover protected reasoning from its models. The 54.0 percent success rate seen on GPT-6 Astra for attacks pulling reasoning from prior requests fell to 0.2 percent on GPT-6.2 Astra, while attacks on active requests fell from 22.9 to 0.9 percent, according to OpenAI. The company says it is now bringing selected improvements to GPT-6 Astra as well.
The defenses operate on two levels. For individual requests, OpenAI combines targeted training, input and output filters for messages and tool calls, and blocks that stop older models from reading protected reasoning. Against coordinated campaigns, it bans thousands to tens of thousands of accounts weekly.
The issue is not unique to OpenAI. Anthropic is fighting systematic distillation attacks as well, also banning tens of thousands of accounts to curb the theft of internal reasoning and reportedly succeeding in 90 to 99 percent of cases. Anthropic has also called for a coordinated response from industry and policymakers, suggesting the companies see this as a shared problem rather than a single-vendor fix.
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.
Home Depot says its Magic Apron AI assistant now fields millions of questions per month and is trained on prod…

Berkshire's $363 billion stock portfolio now holds nearly $112 billion in AI-linked companies, about 31%, unde…

Berkshire Hathaway has been a popular 'anti-AI trade,' but one firm now says it is too expensive, CNBC reports

Applied Materials, which supplies wafer fabrication equipment to Micron and to Nvidia's manufacturing partner…

Anthropic shares fell 8.5% to close at $97.17 on Friday, down from the $111.50 opening price after the company…

Broadcom fell 12% Friday to $306.55 after TF International Securities analyst Ming-Chi Kuo said Google cut abo…
