Claude Code hitting ~50% usage after 1–2 prompts (Pro user) (github.com)

A user of Claude Code, a prominent AI tool, reported experiencing an unexpected spike in usage, reaching nearly 50% of their limit after only two prompts. The pro user noted that this sudden increase was not typical based on their past experience, raising concerns about a potential bug in the latest version of the software. This incident has sparked discussions within the AI/ML community regarding the reliability of usage tracking and the implications it has for users who rely on precise metrics for planning their sessions. This situation highlights the importance of clear and accurate usage statistics in AI tools, particularly for pro users who may face limitations in their access or functionality. As AI applications continue to evolve, ensuring that users have a clear understanding of their resource consumption will be crucial for maintaining trust and usability. If the reported issue persists, it could lead to broader concerns about the tool's reliability and the potential need for adjustments in how usage limits are monitored and communicated.