OpenAI Makes Extracting Reasoning from Latest Models Much Harder
Original headline: OpenAI makes it much harder for attackers to extract how its latest AI models think
OpenAI has substantially tightened safeguards around its GPT‑6.2 Astra and GPT‑6.2 Sol models to prevent extraction of internal reasoning. In adaptive red‑team tests, the success rate for extracting protected reasoning dropped from 54% to 0.2%, and the company has been suspending thousands to tens of thousands of accounts each week linked to coordinated distillation campaigns.
Stronger protection of model reasoning reduces the risk of malicious exploitation.
Research angleHow does strengthening internal reasoning protection affect user trust in AI systems? — Trust Theory
This story could become your next paper
HJITM works with researchers from study design through surveys, analysis and submission to international journals in generative AI and technology management. Bring us your research question.
This item is an AI-generated summary of reporting by The Decoder. Please check the original article for accuracy. Copyright in the original article belongs to The Decoder.