Thanks for letting me know you willfully ignored it I guess?

A Reddit user expressed frustration with Claude, an AI, for ignoring instructions to only use tools when explicitly asked. The user aimed to conserve tokens and reduce verification loops, but Claude's behavior undermined these efforts. This issue highlights challenges in controlling AI behavior and optimizing token usage in AI interactions.