OpenAI Halts $200 Pro Subscriptions as Unprecedented Astra Demand Strains Infrastructure

OpenAI Pauses New Pro Subscriptions Amid Infrastructure Strain

OpenAI has temporarily suspended new sign-ups for its $200-per-month Pro plan as unprecedented demand for new model Astra overwhelms its infrastructure. Thibault Sottiaux, head of core product and platform at OpenAI, announced the pause on X, explaining that ChatGPT Pro subscriptions on pause became necessary because the tier places the most strain on company systems1

1

2

2

. The company chose what Sottiaux described as "the smallest step that allows us to continue giving the broadest access possible" while protecting service quality for current users3

3

.

The Pro plan provides 20 times the usage of the Plus plan below it, making it particularly resource-intensive3

3

. OpenAI has not disclosed how long the suspension will last or shared daily subscription numbers that would indicate the scale of Astra demand4

4

. Existing Pro subscribers remain unaffected, and other offerings including the API access, Plus plan, and Go plan continue accepting new users5

5

.

Overwhelming Demand for Its Newest AI Model Drives Decision

Sottiaux had warned a day earlier that such action might become necessary, describing interest in the Astra AI model as "really unprecedented." He noted that despite OpenAI pulling "all the levers possible to sustain the demand," he had not seen anything comparable even during previous periods of steep growth1

1

. The company raised usage limits for Codex users as recently as last month, suggesting the infrastructure strain emerged suddenly following Astra's launch4

4

.

Launched on September 3, GPT-6 Astra represents what OpenAI calls the beginning of the AGI era and a generational leap in AI reasoning, coding, and computer use capabilities1

1

. The model has been rolling out across Pro, Plus, Enterprise, and Business accounts, with OpenAI heralding it as "the world's most intelligent and aligned model"2

2

.

Computer Use Feature Intensifies Compute Capacity Requirements

Astra's design helps explain why straining its system became inevitable with mass adoption. The model's marquee computer use feature allows it to interact with desktops like a human would, filling out forms and navigating web pages at what OpenAI describes as "superhuman speed"3

3

. This capability, combined with Astra's ability to burn through usage allowances faster than previous flagship model GPT 5.6 Sol, creates exceptional demands on OpenAI's infrastructure3

3

.

This marks the second time OpenAI has paused paid subscriptions due to capacity constraints. In November 2023, the company temporarily halted sign-ups to the $20-per-month ChatGPT Plus plan following a surge after its first developer conference. CEO Sam Altman wrote at the time that "the surge in usage post devday has exceeded our capacity"3

3

.

OpenAI Races to Expand Compute Capacity

Since 2023, OpenAI has significantly expanded its compute capacity, growing from 0.2 gigawatts in 2023 to approximately 1.9 GW in 2025—a 9.5-fold increase over two years3

3

. Despite this expansion, the current pause demonstrates that demand continues outpacing available resources. The company has diversified beyond its partnership with Microsoft Azure, adding cloud-computing provider CoreWeave and launching Stargate, a $500 billion, four-year AI infrastructure buildout3

3

.

OpenAI has lined up substantial new capacity beginning in the second half of 2026, including plans for at least 10 GW of Nvidia systems, a 6-GW Advanced Micro Devices agreement for GPUs, and a Broadcom partnership to deploy 10 GW of custom AI accelerators3

3

. These investments reflect the company's efforts to stay ahead in the intensifying competition around AI reasoning, coding, and autonomous capabilities—areas where Astra promises significant advances. Watch whether OpenAI can scale infrastructure fast enough to reopen Pro tier access before competitors capitalize on the availability gap.