Breathe in Data, Breathe out Poetry?
This is the future of poetry. Data as the birthing ground for grand, extravagant, allegorical verses written by robots. Yes, it is.
Dartmouth’s Neukom Institute for Computational Science will be holding three Turing Tests in Creativity that will benchmark the “humanness” of machine-generated verse, writing and, DJ sets. The competition will be held all through the next academic year and will be open to contestants around the globe.
Love of Poetry & Machine Intelligence
The competition hinges on British computer pioneer and codebreaker Alan Turing’s proposal for a test to determine machine intelligence, i.e. to find out the machine’s ability to exhibit intelligence equivalent, or almost identical to a human. The Turing test, has since been applied to everything from chatbots to imitating Reddit users.
Here’s a summary of the three competitions:
DigiLit will have contestants generate short stories that’ll ideally be discerning with short story selections from the New Yorker or a school’s MFA program. It’ll be similar to NaNoGenMo, a challenge for writing a program that effectively yields an understandable short story. Winner of the most convincing machine-generated short story will get a cash prize of $5,000.
AlgoRhythms will hold a real two-story dance party for judges. At regular intervals, booths will change out for a human or a machine DJ. The DJs will mine from a pre-approved set of 10,000 songs. Those in the audience will be able to choose by phone or polling card whether they think the set was made by a human or a machine. Winner will receive a cash prize of $3,000.
PoetiX is fundamentally the same as DigiLit, however with sonnets rather than short stories. Both DigiLit and PoetiX entries will be judged for humanness and will reward a cash prize of $3,000 to runners-up.
Daniel Rockmore a professor of computer science and director at Neukom said he got the motivation for the challenges while thinking about tracks in a spin class. He believes that the AlgoRhythms had the best chance at succeeding because of the abundance of data officially out there on people’s music tastes. As a matter of fact, there’s already a site created by a 19-year-old developer in Utah named Derek Dapp, Actually, we’ve as of now got a site – isitabanger.com – that can let you know whether a tune’s a banger, or not. The other two competitions are more complicated, since one has to code syntax, cadence, mood, etc. so that a machine can identify what makes a composition work.
Prof. Rockmore believes that in the future, an artificial being even fall in love. However, what was fascinating to him was that robots didn’t take up every job that humans do— they couldn’t compose love letters the way the love smitten people do.
Perhaps, it will happen someday, when machines acquire that aesthetic sense that we Earthlings do.