OpenAI pauses $200 Pro subscriptions as Astra demand strains infrastructure

OpenAI has temporarily stopped accepting new sign-ups for its $200-per-month ChatGPT Pro plan because demand for Astra, its newly released flagship model, is overwhelming the company's infrastructure. According to TechCrunch, the suspension was announced on X by Thibault (Tibo) Sottiaux, OpenAI's product lead for core products including ChatGPT and Codex.

What changed

Sottiaux said the Pro tier places the heaviest load on OpenAI's systems of any consumer plan, which is why new subscriptions for that level have been switched off while other options stay open. The company "wanted to take the smallest step that allows us to continue giving the broadest access possible," he wrote, according to TechCrunch. Existing Pro subscribers are unaffected, and the lower-cost Go and Plus plans plus API access remain available to new customers.

The pause had been telegraphed. On Wednesday, Sottiaux wrote that demand for Astra was "really unprecedented" and that OpenAI was "pulling all the levers possible" to sustain it, adding that he had not seen anything like it despite having lived through very steep growth before. He said keeping service quality high for existing users was the priority, and warned that new Pro sign-ups might have to stop for a while if the surge continued.

OpenAI has not said how long the pause will last, and it has not disclosed how many people are signing up each day, so the true scale of the spike is difficult to judge from the outside.

Astra's rapid rollout

Astra launched on September 3 and has been rolling out across Pro, Plus, Enterprise and Business accounts. According to TechCrunch, the model is positioned as a major advance in reasoning, coding and computer use — three areas of steep competition among AI labs. OpenAI has also promoted it aggressively, framing it as the start of the "AGI era" and a generational leap, messaging that has likely fed the surge in interest.

The capacity crunch appears to be a recent phenomenon. As TechCrunch points out, OpenAI raised usage limits for Codex users as recently as last month, which suggests the infrastructure was holding up until Astra arrived and bent the growth curve upward.

Why it matters

The episode is a reminder that AI demand is still gated by physical infrastructure, not just model quality. OpenAI is effectively turning away revenue from its priciest consumer tier because it cannot guarantee capacity for new customers — a striking trade-off for a company whose business model depends on scaling usage.

It also illustrates the double-edged nature of high-stakes launches. By billing Astra as the beginning of an "AGI era," OpenAI generated enthusiasm that its own systems are now struggling to absorb, forcing it to choose between new income and service quality for the users it already has.

For everyone else watching the reasoning, coding and agent-capability race, the pause is a signal that even the best-funded AI lab can hit hard capacity ceilings when a model lands well — and that reliability, not just raw capability, may become a real differentiator as these systems move into everyday work.