A meat proxy forwards agent output to a colleague unread, so the receiver inherits the verification the sender skipped and cannot tell what was checked.
It’s the cadence, it’s very Claude. Also looking at it again on a desktop browser the sidebar shows endless LLM-related posts and the footer contains “Built with AI, for practitioners of AI.”
This is a fairly obviously LLM-authored post. Surprised to see it get upvotes here.
Other than em dashes (and not used how LLMs typically use them), what makes you so sure this wasn’t written by a human?
It’s the cadence, it’s very Claude. Also looking at it again on a desktop browser the sidebar shows endless LLM-related posts and the footer contains “Built with AI, for practitioners of AI.”