New Yorker article on the "Machine Writing": newyorker.com/magazine/2019/… #ai
In this strange new world enabled by "tab", our writing may end up to be composed mostly of emojis.
"The creepy thing was that the machine was more thoughtful than I was"
"the carbon footprint created by training a gigantic neural net is roughly equal to the lifetime emissions of five automobiles."
“It’s like alchemy! ”
"writing seems to be a whole-brain activity—a brainstorm indeed."
"Today’s A.I. systems are good at one or the other, but it’s hard for them to put the two kinds of learning together the way brains do."
"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."
"The mathematical calculations that resulted in the algorithmic settings that yielded GPT-2’s words are far too complex for our brains to understand."
"It hurt to see the rules of grammar and usage, which I have lived my writing life by, mastered by an idiot savant that used math for words. "
“I’m betting yes. That’s what cognition is all about, a hybrid architecture that combines different classes of thinking.”
"Humans would stop writing, or at least publishing, because all the readers would be captivated by the machines. What then?"
It's always very interesting when a human expert, in this case a writer from the New Yorker, encounters a deep learning system. In just a few years, we've had reactions from image processing researchers, NLP researchers, Go players, chess players and protein folding experts.
@threadreaderapp unroll
