Automation, Algorithms, And Android News Readers: How Robots Are Changing The Face Of Journalism
It's happening already.
The year is 1986, and radio listeners in Adelaide are shocked. ABC Local Radio has just announced that its newsreaders are being replaced by advanced voice synthesis robots. The station blames reports that even the tone of someone’s voice could introduce bias to a neutrally written story. Veteran radio newsreader Alf Jarvis, fresh from announcing his retirement, rails against this change and declares he is glad to be getting out of the industry. A spokesman for the public sector union also voices their concern for job losses.
But the date is April 1.
Technology writer and broadcaster Stilgherrian fondly recalls his prank: “I was one of the producers working on the 8:30am daily talk show. That morning we ran the story straight after AM, and the talkback board lit up with outraged callers. We even used one of the early Mac computer programs to read the day’s news. It was great fun.”
Fast forward thirty years, and robot journalists are closer to becoming a reality. In June this year, Osaka University Professor Hiroshi Ishiguro unveiled Kodomoroid: the world’s first news-reading android. After years of research and development it can pronounce complex tongue twisters, speak multiple languages and interact with people. It looks like this:
Professor Ishiguro is globally recognised for his contributions to robotics; he’s even created a robot version of himself, that he sends around the world to deliver lectures on his behalf. His latest creation has silicon skin and artificial muscles, making it look just like a human. Powered by compressed air and servomotors, it replicates natural movements including blinking and twitching its eyebrows.
During its demonstration, the robot read the news with perfect language skills, and joked with the crowd that one day it would like to have its own television show.
If current trends continue, this dream isn’t so far off.
The Rise Of Robo-Journalism
Robot journalism has come a long way since the first story-writing algorithm was developed at Yale University 40 years ago. Many people are already reading stories written by robots and might not even realise it, unless the outlet is transparent — as was the case in March last year, when the Los Angeles Times published a news report on an earthquake aftershock that was created by an algorithm written by the author.
American-based company Automated Insights is pioneering this new technology; their Wordsmith program searches raw data for information and trends, and adds context to form a narrative. But the software doesn’t just fill in the blanks — its advanced algorithms are able to add perspective, tone, and humour to a story. Their many clients include Samsung and Yahoo!, for whom they provide over 50 million personalised fantasy football reports.
Here is a test from the New York Times: guess which piece was written by a robot, and which by a human.
“Things looked bleak for the Angels when they trailed by two runs in the ninth inning, but Los Angeles recovered thanks to a key single from Vladimir Guerrero to pull out a 7-6 victor…”
— New York Times, 7th March, 2015
–
“The University of Michigan baseball team used a four-run fifth inning to salvage the final game in its three-game weekend series with Iowa, winning 7-5 on Saturday afternoon (April 24)”
— New York Times, 7th March, 2015
The first one was a robot.
The technology is not just being used for sports reporting. The world’s oldest newsgathering organisation, the Associated Press, started using Wordsmith to write their quarterly earnings reports in 2014. Since then, they’ve published ten times as many reports — encouraging them to rely on Wordsmith for college sport reporting, too.
In the early days, humans checked every piece generated by the program, with errors reported to Automated Insights. But since October of last year, the AP has published stories without human verification. Its popularity is rising, with the Poynter Institute estimating that last year alone Wordsmith produced more than one billion pieces of individual content.
What Does This Mean For Human Journalists?
The rise of machines is often associated with the loss of human jobs. And with large Australian media companies like Fairfax continuing to divest in their regional holdings, it’s not a huge stretch of the pessimist’s imagination to predict that — as budgets get tighter, and these programs get better — journalists in these areas could be replaced by programs like Wordsmith.
But Professor of Media at Macquarie University John Potts urges for everyone to remain calm; he believes there’ll always be a place for people in good quality journalism. “There’s only so far it can go. Journalism requires judgment which is a human quality based on personal experience and other faculties,” he says. “Journalists must be able to judge when to push for information, be able to frame stories, and be aware of ethical constraints. None of these processes can be automated.”
It makes sense: so far, the stories currently being written by programs like Wordsmith are purely informational, crafted by parsing data and narrating it through an accessible sentence-structure. Rather than hurting a journalist, they can actually help: a computer-generated story describing the fall of the Australian dollar could help a journalist analyse what effect it will have on the average Australian family’s living expenses. Robot journalists could well become research assistants for their human counterparts; by using computers to find and order information, all a journalist would have to do is add their own interpretation, saving them time and making them better value for money.
Looking Ahead
Computer generated news programs can make sense of massive amounts of data in a flash; at its peak, Wordsmith can produce 2000 articles in under a second. As media companies face the challenges of declining subscription and advertising revenue and a crowded marketplace, the program could help them cut costs while increasing their output in a saturated online environment.
These developments could also push towards more personalisation in news. The logical extension of these algorithms is that they’ll use data gathered from your smartphone or your Apple Watch to produce unique stories for you. For example, instead of speculating about how much a new road toll might cost motorists, these services will know your travelling habits and deliver a report that tells you exactly how much more you will have to pay per year.
For better or worse, automation will have a place in the future of journalism. When asked if he would be happy to wake up and listen to news that is written and presented by robots, Stilgherrian echoes a popular sentiment: “It depends on how good the robot is.”
–