Report
OpenAI reportedly accuses a Chinese AI lab of trying to extract hidden model reasoning
A user says the alleged operators copied encrypted reasoning from one conversation and asked another model to decrypt and transcribe it, rather than breaking encryption.
TLDR
A user says OpenAI accused a Chinese AI lab of trying to extract its models’ hidden reasoning. The account describes operators copying encrypted reasoning between conversations and cites 16,000 requests from over 4,000 users during July 24–25 spikes. It says OpenAI disrupted related activity across more than 15,000 users by July 28, closed the replay pathway and banned accounts.
Combined views
2.7K
1 Source, first seen 7h ago
