They're not making the argument that you can't get better results if you design a prompt just right. They're making the argument that it's "not a reliable inter

They're not making the argument that you can't get better results if you design a prompt just right. They're making the argument that it's "not a reliable interpreter of the world." It's still completely on humans to guide it, to work around the limitations that come from the algorithm not knowing what words or sentences mean. In that sense it's similar to the mechanical turk with a thin but impressive layer of automation that does a neat trick but not what's ultimately the important part of communication.