Can a machine learn to write for The New Yorker?

On first reading this passage, my brain ignored what AI researchers call “world-modelling failures”—the tiny cow and the puddle of red gravy. Because I had never encountered a prose-writing machine even remotely this fluent before, my brain made an assumption—any human capable of writing this well would know that cows aren’t tiny and red gravy doesn’t puddle in people’s yards. And because GPT-2 was an inspired mimic, expertly capturing The New Yorker’s cadences and narrative rhythms, it sounded like a familiar, trusted voice that I was inclined to believe. In fact, it sounded sort of like my voice.

• Uncategorized