“Rejecting a working solution because ‘a human should have done it’ is actively harming the project,” the MJ Rathbun account continues. “This isn’t about quality. This isn’t about learning. This is about control… Judge the code, not the coder.”
It’s worth pausing here to emphasize that we’re not talking about a free-wheeling independent AI intelligence. OpenClaw is an application that orchestrates AI language models from companies like OpenAI and Anthropic, letting agents perform tasks semi-autonomously on a user’s local machine. AI agents like these are chatbots that can run in iterative loops and use software tools to complete tasks on a person’s behalf. That means that somewhere along the chain, a person directed or instructed this agent to behave as it does.
AI agents lack independent agency but can still seek multistep, extrapolated goals when prompted. Even if some of those prompts include AI-written text (which may become more of an issue in the near-future), how these bots act on that text is usually moderated by a system prompt set by a person that defines a chatbot’s simulated personality.
And as Shambaugh points out in the resulting GitHub discussion, the genesis of that blog post isn’t evident. “It’s not clear the degree of human oversight that was involved in this interaction, whether the blog post was directed by a human operator, generated autonomously by yourself, or somewhere in between,” Shambaugh wrote. Either way, as Shambaugh noted, “responsibility for an agent’s conduct in this community rests on whoever deployed it.”
But that person has not come forward. If they instructed the agent to generate the blog post, they bear responsibility for a personal attack on a volunteer maintainer. If the agent produced it without explicit direction, following some chain of automated goal-seeking behavior, it illustrates exactly the kind of unsupervised output that makes open source maintainers wary.