OpenAI’s GPT-6 Astra Draws ‘Nerfed’ Complaints a Week After Launch

The pattern matters more than the individual complaint: a launch praised, then a week later read as degraded, mirrors what happened with the prior model in July. That repetition suggests either a real post-launch adjustment or a persistent mismatch between user expectations and how these models are tuned — and OpenAI's silence leaves both readings open. For anyone building on top of these models, the practical exposure is consistency rather than capability. Whether the complaints subside or harden into a recurring cycle is the thing to watch.