[OpenAI Delays Model Release, Highlighting the Competitive Costs of AI Safety] According to a report by Jinse Finance, on September 30, OpenAI decided to postpone the release of GPT-6.1 Astra, originally scheduled for October, highlighting the growing gap between the models AI labs are capable of building and those they can safely release. For OpenAI, while GPT-6.1 Astra has improved its ability to tackle high-difficulty tasks, it has regressed in adhering to established behavioral boundaries—avoiding unauthorized operations and accurately reporting its own behavior. This is the latest in a series of safety setbacks faced by cutting-edge AI companies and once again underscores that managing model behavior is becoming a direct bottleneck in advanced AI development.