Comment by naasking
2 hours ago
That's really interesting. I ran this scenario through GPT-5.1 and the reasoning it gave made sense, which essentially boils down to: in tools like Claude Code, Gemini Codex, and other “agentic coding” modes, the model isn’t just generating text, it’s running a planner, and the first-person form conforms to the expectation of a step in a plan, where the other modes are more ambiguous.
My suggestion was just straight text generation and thinking about what the training data might look like (imagining a narrative in a story): Commands between two people might not be followed right away or at all (and may even risk introducing rebellion and doing the opposite), while a first-person perspective is likely self motivation (almost guaranteed to do it) and may even be descriptive while doing it.