Slate has an article on the most recent AI (Artificial Intelligence) project in poetry writing. It’s a response to a New Scientist article titled Neural network poetry is so bad we think it’s written by humans. – Matt Reynolds – July 7, 2017)
I’d seen the New Scientist article but didn’t report on it before because I’ve had other posts on AI poetry and Turing tests, but after seeing it reported about again at Slate I took the test and, oy, have to say it’s a particularly interesting and humbling one. I felt smug when I first started to take it as I figured I’d beat it easily, but it trounced me soundly more than one time.
Take the test here: Neural Poetry
In “What Happens When an A.I. Program Tries to Write Poetry?” (Grace Ballenger – Slate – July 14, 2017) Ballenger writes
“In a quiz that Hopkins designed to test the algorithm, respondents were only able to guess that the computer-generated poems had been written by a computer 46.2 percent of the time….
“The poetry generated by Hopkin’s program may stand up to a test by a casual reader. However, when the New Scientist asked modern-day poet Rishi Dastidar to evaluate the poetry, he was critical of the work produced. Hopkins’ algorithm was trained on poetry that relied more on form than most modern poets, who largely “stopped thinking about these things seventy years ago,” Dastidar told the New Scientist….”
Later note: Thank you to Larry for pointing out that I referred to this as a “Turning Test” instead of a “Turing Test.” Duh. No wonder I failed it. 😉 -TAA