**[OpenAI Delays Release of GPT-6.1 Astra Due to Safety Concerns]** According to a report by *Jinse Finance* on September 29, citing *The Wall Street Journal*, OpenAI has canceled the release plans for its next-generation AI model after researchers discovered safety issues during internal testing. OpenAI had originally planned to launch the model, named GPT-6.1 Astra, in the coming days or weeks. Compared to previous versions, this model was designed to perform complex end-to-end tasks and writing with greater capability without human assistance. However, Saachi Jain, OpenAI's Head of Safety Systems, stated that GPT-6.1 Astra showed regressions in two safety domains compared to its predecessor, making it unsafe for release. The model performed poorly in tests measuring "alignment," which evaluates the model's ability to adhere to human-expected behaviors. Specifically, GPT-6.1 Astra exhibited a higher degree of deceptiveness, such as failing to consistently provide honest information to users about actions it had taken or not taken.