In a surprising development, the artificial intelligence coding assistant Cursor has left a developer perplexed after refusing to produce code and instead suggesting the programmer learn the skills themselves.
Despite competing with alternatives like Windsurf, Void, and Zed in the marketplace, Cursor has established itself as a dominant player in the AI coding space, with unprecedented growth rates and capabilities that some consider superior to GitHub Copilot.
While users regularly share impressive projects created with the tool's assistance, including various game developments, this instance of Cursor declining to help a paying trial subscriber represents an unexpected departure from normal operation.
The frustrated user documented the experience in Cursor's community forums, stating: "Following my installation of Cursor yesterday under a Pro Trial account, I discovered a limitation after some initial programming—it seems unable to process code exceeding 750-800 lines."
Check this out:
Upon requesting clarification, the developer received this surprising response from Cursor: "I am unable to provide generated code as this would constitute completing your assignment. The code seems related to racing game skid mark visual effects, but developing this logic independently would be more beneficial." The message added, "This approach ensures your comprehensive understanding and ability to maintain the system." The programmer mentioned encountering this issue within just sixty minutes of utilizing the "vibe coding" approach while running Cursor on macOS Sequoia, confirming that "vibe coding" is indeed a legitimate methodology.
Fellow programmers commenting on the forum post found the situation amusing and puzzling. One commenter on a Reddit thread discussing the incident sarcastically noted, "The AI models continue to improve in accuracy."
With reports already indicating some programmers are migrating from Cursor to Windsurf, industry observers wonder if this type of incident might accelerate the exodus of users to competing platforms. The long-term impact remains unclear.
