Machines write the news

Last week, a robot wrote a breaking story. But it wasn’t the first JUST 3 minutes after an earthquake hit California last week, the Los Angeles Times broke the story. The short web article seemed ordinary. It covered all the major details – when the quake hit, its magnitude and how far it spread. The only sign of anything unusual was the final sentence, which ended with: “this post was created by an algorithm written by the author”. In other words, the article was put together by a robot. Once readers realised the story was computer-generated, it attracted a lot of attention. But quite a few machines already write the news. Forbes magazine uses a company called Narrative Science, in Chicago, to robotically report on corporate earnings. And Narrative Science started working with ProPublica in January to create  an application to write summaries of 52,000 schools in a database. The earthquake algorithm, nicknamed Quakebot, isn’t the only robot reporter used by the Los Angeles Times. The newspaper’s Mapping LA project has bots to compare neighbourhoods, and its website, The Homicide Report, automates posts about murders in the city. Although robot reporters cannot yet file fascinating 2000-word features for New Scientist, early indications of their skills are promising. A study last month in Journalism Practice found that none of a group of 30 readers could reliably discern whether a sports article was written by a human or a bot (doi.org/r2g). Those reading the automated article found it trustworthy and informative, albeit a bit boring. “Would this replace real reporters? I would say no,” says Christer Clerwall of Karlstad University in Sweden, the study’s author. He believes that robots will continue to be given mundane, data-aggregating stories, leaving more complex tasks like narrative and eyewitness reporting to humans. But robot reporters have plenty of benefits, says Quakebot creator Ken Schwenke. “We had it up and we had it first and we had the information out for people,” he says. “If we can automate it, why not?” Just like statistician Nate Silver, who famously predicted the last US election result, robot journalists are all about data. They write stories by crunching spreadsheets full of sports SUZANNE PLUNKETT/REUTERS scores, sales numbers or stock market fluctuations. When it comes to pure computing power, bots will beat humans every time. Quakebot works by filling in gaps. When the US Geological Survey sends an email alert about an earthquake, Schwenke’s bot parses the email, slots the data into a template and uploads the article to the newspaper’s content management system. It even sends an email to tell the editors to look it over. Other approaches are more complex. Automated Insights in Durham, North Carolina, builds robot reporters that “We had it up first and we had the information out for people. If we can automate it, why not?” look for interesting trends. They scour data to focus on stories that are of interest to only a small audience –  like recaps of how someone’s fantasy football team has performed, or summaries of recent web statistics. In 2013, the company churned out 300 million pieces of content. Most journalists, says CEO Robbie Allen, want to write one article that will be read by lots of people. Automated Insight’s goal is to do the opposite. “We’ll create a million pieces of content that we hope a million people read just one of,” he says.  Aviva Rutkin  ■