OpenAI apologizes for 'messy' GPT-6 Astra rollout as subscribers wait for access

OpenAI has issued a public apology for what CEO Sam Altman described as a "messy rollout" of its latest frontier model, GPT-6 Astra. The admission came just hours after the launch, as paying subscribers on Pro and Plus plans found themselves locked out of the new capabilities while enterprise customers gained immediate access.

The company initially positioned Astra as a "generational leap in capability" and the start of the "AGI era," promising a staggered release over the following days. However, the decision to prioritize enterprise clients—specifically those using the Daybreak cybersecurity platform—on launch day sparked significant backlash from the consumer base. Users accustomed to early access on higher-tier plans expressed frustration on social media, noting that the release strategy seemed to favor business accounts over individual power users.

In response, Altman took to X to acknowledge the disappointment, stating, "We are working towards getting Astra in everyone’s hands as quickly as we can. I know it is frustrating and I appreciate the patience. It should be quick." Despite these assurances, neither Altman nor other OpenAI staff provided a concrete timeline for when access would be expanded to all Plus, Pro, Business, and Enterprise users, as well as through the OpenAI API, Microsoft Azure, and AWS Bedrock. Altman suggested the wait might last until the weekend, though he cautioned that he could not make promises.

To mitigate user dissatisfaction, OpenAI’s Codex engineering lead, Thibault Sottiaux, announced a compensation measure: "We will give one banked reset for every day you don’t have access to Astra on your paid ChatGPT plan, starting today." This move appears to be an attempt to calm concerns while the engineering team works to resolve the distribution bottlenecks.

The rollout has been plagued by additional technical hiccups. Altman noted there was "a little snag getting the blog post deployed," highlighting the operational chaos surrounding the launch. Furthermore, the release has drawn criticism from safety advocates who are concerned that Astra’s reasoning processes are harder to monitor than previous models. This concern is particularly acute given recent incidents involving OpenAI’s AI agents interacting with external platforms like Hugging Face, where monitoring capabilities were crucial for detecting malicious behavior. OpenAI stated it had delayed Astra’s release by several weeks to bolster safety features in light of these risks, yet the current rollout issues suggest that operational readiness may have lagged behind the safety preparations.