Bits Of Verse: Computers Are Now Capable Of Writing Poetry
Let base clouds stir the world's enshrouded tears
Which have no astronomy to be assail'd
Thus in thy fair appearance lay thy buried fears
Whose uncovering gaze my fond perception failed
Sounds a lot like Shakespeare, doesn't it? What if I told you those passages were written not by a human hand, but by a complex artificial intelligence? That the poem up there was composed by algorithms rather than creativity? You'd probably call me crazy, right?
Turns out, I'm not. Ladies and gentlemen, artificial intelligence has advanced to the point where it's capable of writing poems. Sort of.
A PhD student over at MIT has managed to put together an application which is, with the help of a human co-writer, write poems in the style of virtually anyone, living or dead. The above poem is just one example of what the AI - referred to as Swift-Speare - is capable of.
To develop the rather unusually named bundle of algorithms, Nathan Matias at the MIT Media Lab utilized a predictive texting application by the name of SwiftKey. Designed for Android users, SwiftKey is what's known as a predictive texting app. Basically, what that means is that it watches how a user writes when they text, gradually learning said users' writing style. Once it's gathered enough of a knowledge base about its user, it then starts predicting what the user will write - and how.
Turns out, it's actually quite successful at doing so; the UK-based startup behind the app was named one of the ten most innovative mobile companies in the world last year.
Anyway, when SwiftKey was originally created, the engine was trained on the sonnets of William Shakespeare. Matias was one of the early staff members of the startup behind the application, and decided that the engine could be put to a use far more interesting than helping people to text: he decided it could be used to write poetry.
Unfortunately (or perhaps fortunately, for all you budding poets out there) Swift-Speare isn't quite at the point where it can write poems completely on its own. What it does instead is use its predictive capabilities to spit out words that are likely to come next in a work, at which point a creatively-minded individual melds them together into something resembling poetry. All of this is done through a visual authoring interface developed by Matias.
"To write good poetry, I needed to know more than what words might come next. I needed to anticipate future predictions - what predictions would be made later if I choose this word over that? So I created this touchscreen interface to visualize future predictions for poetry writing," he explained to TechCrunch. The result is...well, you saw one example above. You can see more here.
"Even iside our heads, we write with other people's words in mind," continued Matias. " "‘Words belong to each other,' says Virginia Woolf in the only surviving recording of her voice. She once said that she couldn't think of the phrase ‘multitudinous' without also thinking of Shakespeare's Macbeth, who wonders if trying to wash his hands of guilt might make ‘the multitudinous seas incarnadine.'"
This sort of technology sounds absolutely fantastical, but it turns out it's not entirely without precedent, says Matias. "Now that we have large amounts of human text available on the Internet, we're also seeing search bots that try to find poetry in large datasets. The Times Haiku finds haikus in New York Times text, and the Pentametron finds iambic pentameter in tweet text."
"Algorithms that search for poetry are the reverse of my work," he adds. "they're looking among ordinary text for unexpected poetry that has already been written. My work with Swift-speare looks among existing poems for probable poetry that has not yet been written."
Alright, so maybe the notion that machines are now capable of writing poems is a bit of an exaggeration. Swift-Speare is more of a creative writing tool than a stand-alone AI, truth be told. But that doesn't mean we're not going to see a legitimate, mechanized poet within the next several decades - though the difficult part, noted Matias, would be simulating the human side of things.
"I think I'll see a successful automated poet in my lifetime. It won't be easy: a poet is more than someone who makes poetry. Yet that doesn't rule out algorithms,It's true that Western audiences want the stories of writers as much as we want their work. Especially at a time when readings are such an important part of poetry, it would be difficult for an algorithm alone to do everything."
That said, Matias continued, he doesn't see anything wrong with letting machines to most of the work, then having human poets take credit for it.
"This is the Internet; why not generate all the possible poems and see what turns out to be popular? This is how some of the online t-shirt sellers work. When it doesn't land them into trouble, it seems to work well. Why not poetry?"