One user's frustration with Cursor, the AI code editor, went public this week after they published a detailed GitHub report claiming the tool caused their system's RAM to climb to 61GB during a normal coding session.
To put that number in context: 61GB of RAM would exhaust the memory on almost any developer laptop. High-end consumer machines ship with 16-32GB, and even workstations rarely go higher. A memory spike that size wouldn't just slow things down - it would crash most machines entirely.
The more serious allegation follows from there. After raising the issue on Cursor's community forums, the user claims their posts were shadowbanned - visible to themselves but hidden from other users. They describe the outcome as deliberate suppression of a legitimate bug report.
This is a single-user account with no independent verification, and Cursor has not publicly responded. The shadowbanning claim is especially hard to confirm from the outside.
Memory problems in AI code editors aren't implausible: these tools maintain persistent connections to cloud AI models, build and index local codebases, and run real-time completions in the background - a combination that can accumulate fast if a memory leak goes unchecked. But 61GB is an extreme figure that would need reproducible evidence before drawing firm conclusions.
The GitHub report includes screenshots and a full timeline for those who want to evaluate the claims directly. Whether or not the forum moderation happened as described, the resource usage complaint deserves a public response from Cursor.