Robots are getting better at writing content. Here's why that's good news for journalism
What do we talk about when we talk about "content"?
We say that it's "king," we say that it's a smart investment (though maybe not for long). We speak of the "Content Wars," wherein dozens of news outlets rewrite the same viral stories about toilet-paper-eating moms or a fancy new Star Wars trailer in a mad dash for web traffic. The victors of the Content Wars are determined not by the "content" itself, but by clickable headlines, blind luck, and Facebook's algorithms.
But is this "journalism"? Sam Slaughter -- who, as the VP of Content at the content marketing firm Contently, knows a thing or too about the big C-word -- told me, “There’s this difference between ‘journalism’ and ‘content.'”
What he means is that the primary purpose of content is to entertain or educate -- like when an outlet reposts a fun viral video it found on Reddit, or creates an "explainer" on the best way to remove a ketchup stain. Unlike journalism, these stories aren't about power dynamics or corrupt institutions or nuanced problems. And therefore, because it's only content not journalism, there's nothing wrong with a brand producing it.
I would take it one step further: There's nothing wrong with a robot producing it either.
And so it goes with "Content, Forever," a new algorithm developed by self-proclaimed "Internet artist" Darius Kazemi. Here's how it works: You enter in the number of minutes you have to consume the content, followed by a topic. Hit "submit," and within seconds you'll receive a personalized piece of content, hot and fresh out the kitchen, for you to consume.
The algorithm isn't perfect -- for example, when I searched for a minute-long piece on "Drake," I got back a bizarre, free associative rant that jumped from the recording artist Drake's self-released mixtape, to a short treatise on the art of self-publishing, to a history of Drake's hometown of Toronto, Canada.
Nevertheless, it's a yet another step toward robot-journalists -- a somewhat scary thought for a field that always considered itself impervious to automation. But in the same way that brands are producing more and more of their own content, who cares if a robot writes a short post about a viral puppy video? Do we really want to waste human hours on such a thankless task -- hours that could be better spent poring through public documents or interviewing sources -- in essence, doing the things that brands shouldn't do and that (for now) robots can't do?
Here's a possible model for the future of media: The bulk of the grunt work -- and let's be honest, that's what most "content" is -- can be outsourced to brands as "native advertising," thus creating more revenue for the news organization. All the content that can't find a sponsor will be created by robots, allowing the rest of the workforce to concentrate its resources on real reporting and journalism. This model would work especially well for aggregation. After all, what is aggregation if not reading someone else's story and then rewriting it by moving and replacing some words to avoid plagiarism? Sounds like a job uniquely tailored to a robot's skillset.
And that's why I'm not afraid of robots taking food off of journalists' tables. But as for content hacks who trawl the web looking for trending stories, only to rewrite them and watch the traffic roll in? They are officially on watch.
[Photo credit: Roby Ferrari (Creative Commons)]