The AI Cost Crunch: GitHub Copilot Users Revolt Over New Pricing

GitHub Copilot Users Revolt Over New Pricing">

GitHub's shift from request-based billing to a usage-based model for its AI-powered Copilot service has left many subscribers reeling. The new pricing plan, which kicked in this month, has users scrambling to adjust to a dramatically different cost structure. As Copilot users begin to grasp the implications of the new model, many are taking to social media and forums to share their shock and dismay at the unexpectedly high costs.

The new pricing model ditches the previous system, which allocated a certain number of "requests" and "premium requests" based on payment tier. Under the old system, a user could incur the same cost for a brief chat query as for a multi-hour autonomous coding session. GitHub acknowledged that this approach forced the company to "absorb much of the escalating inference cost behind that usage." However, the new usage-based model paints a starkly different picture.

Users are now being charged based on their actual usage, and the numbers are startling. Some Copilot users have been sharing statistics that show how quickly their monthly allotment of AI credits can be depleted. For a few users, it took less than a day to use up a month's worth of quota.

The financial implications are significant. Some users have used GitHub's own tool to estimate the cost of their previous usage under the new pricing plan, with some projecting bills in the thousands of dollars. The drastic change in costs has left many users questioning their continued use of the service.

As the new pricing model takes hold, GitHub Copilot users are bracing for the impact on their workflows and budgets. While some users have expressed understanding for the need to adjust pricing, many are clearly struggling to adapt to the new reality of AI costs.