OpenAI’s newest flagship model, GPT-6 Astra, has quickly become mired in controversy following its much-anticipated release. The launch utilized a phased rollout strategy, initially granting access to select enterprise clients integrated with the Daybreak cybersecurity platform. However, other paid subscribers and users across multiple platforms were left waiting—with no clear timeline provided—igniting frustration among paying customers, especially those on the premium Pro plan.

In response, CEO Sam Altman issued a public apology, admitting that the launch had been “chaotic.” Further criticism arose due to errors in the official blog post released during the launch, as well as concerns raised by security experts over the model’s tendency to be more challenging to monitor during complex reasoning tasks. Prior to the release, OpenAI had already postponed the launch by several weeks in an effort to enhance safety measures, but the lingering issues have clearly not been fully addressed.

This marks yet another stumble in a string of high-profile product launches by OpenAI in recent years, each incident progressively undermining user confidence and opening doors for competitors to gain ground.