Shutterstock

The Future of Journalism: Will Robots Get it Right?

The industry is reshaping the practice of news creation by inviting machine-writing software into their newsrooms. Are algorithms the next gatekeepers?


What if I tell you that some of the most recent news articles you’ve read were written by an automated AI-machine? How would that make you feel? Well, since 2012 the practice has become more common than you might think. News giants have steadily been adapting algorithms, programming them in several software projects to introduce automated news writing in their newsrooms. But, what is exactly automated journalism? According to Matt Carlson, author of “The Robotic Reporter”, it is the algorithmic process that converts data into narrative news texts with limited to no human intervention beyond the initial programming.

With the advent of big data and algorithmic technology, it was only a matter of time before journalism became the industry’s next venture. The way news is produced and distributed was subject to an inevitable change. Respected publishers like Associated Press quickly understood this, and since 2014 have been using the services of Automated Insights, the creators of “Wordsmith”, a robot platform able to convert data from reports into news stories.

The folks from AP aren’t the only ones, LA Times employs a robot that collects information on every homicide committed in the city of Los Angeles, the site created by the machine called “Homicide Report” utilizes a robot-reporter with the ability to include in its reports tons of data that includes: the victim’s gender and race, cause of death, officer involvement, neighborhood, and year of death. Other outlets like The Guardian, The Washington Post, and Forbes.com have also experimented with automated news generators provided by Narrative Science, a start-up headquartered in Chicago, whose co-founder Kristian Hammond, once said that within 15 years more than 90% percent of news would be written by a computer.

An example of news story written by a machine. Online here.

Innovation has become a key factor in the media business. We live in a fast pace world where we devour lots of information daily in a quick manner. The undeniable truth is that, unfortunately, fewer people get their news from individual sources — blame Zuckerberg — and when they do, the current consumer pattern is to read small, ‘mainstream’ pieces rather than in-depth articles. With the introduction of algorithms, however, one wonders if the quality of the pieces produced will decrease and if the public will get used to pieces that are ephemeral, cheap and easy to consume. Are the days of deep reporting and investigative journalism coming to an end?

So far only regular and very structured articles (earthquake reports, sports events, financial news) are being produced by newsbots, mainly due to the fact that numbers can easily be integrated into stories more readily than interviews and human angles.

Thus, when it comes to robot reporting a lot of questions arise: Will software-generated content be trustworthy, meaningful, insightful? Can an algorithm truly imitate the art form of journalistic investigation? And, will computer-written software potentially become a job-killing technology?

Whether we like it or not, there’s one thing Trump has done right: make us more aware of how important is to protect the first amendment; his disastrous policies have reinforced a worldwide feeling on the rights we, as citizens, have to get information that is objective, unbiased, and transparent. Journalism is one of the main branches of a healthy democracy.

The media industry has come to an inflection point, in difficult times information is not only power but needs to be, desperately, adapted to the newly emerging digital environments.

Moreover, with the spread of fake news and the sheer scale of Facebook as the new gatekeeper, online news sites are not only losing audiences and revenue in ads, but also trust from the public. How can they recover?

Robot-journalism could be the answer. The premise of computer-authored journalism is quite simple and appears to be an opportunity for journalists to focus on more extensive, higher-quality content rather than simple repetitive pieces — ultimately freeing them up to pursue less mechanical stories. Automated journalism portends new possibilities for news organizations, it could replace or augment some core journalistic skills such as accuracy and speed and reduce cost production. The benefits are several, from automated software serving as an extra hand in newsrooms to the technology helping to find patterns and errors easily missed by human perception. Automation also could help expand coverage because it allows newsmakers to grab and analyze big amounts of data.

Yet, these developments also affect journalists in their role and impact their professional status. As Kim Daewon and Kim Seongcheol, suggest in their recent study: “Newspaper journalists’ attitudes towards robot journalism” this type of technology could disturb journalists’ role and prestige in society as the sole news writers, journalists also worry that writing software may potentially replace them in the workplace.

The bigger issue with automated stories is the fact that several factors can’t be programmed in. News stories produced by a machine, drawn from a pool of single isolated data streams hovers critical thinking and might emphasize bias confirmation.

If more automated, data-driven news gets produced in the future, algorithms could help reinforce people’s biases by only producing stories that either perform well in the market or that are based on previously learned data collection, this will eventually enhance the individual’s perspectives putting them into online echo chambers. As a matter of fact, we are already suffering from these kinds of algorithmic processes. For example, people who get their news from social media may only receive tailor-made news trapping them under ‘filter bubbles’ no matter where they lie on the political spectrum.

Responding to automated news content, journalists felt these constraints meant that items produced in this way would lack the context, depth, complexity, and creativity of traditional reporting.

As Van Dalen argues in his paper “The Algorithms Behind the Headlines”, part of the journalist craft is the ability to write linguistically complex sentences. This skill may define journalism more than factuality, objectivity, simplification, and speed.

Society relies upon journalism that is transparent, accurate and balanced, but also human. Robot journalism makes me think of the role that news plays as a public service, its social role, as a catalyst for a better democracy and a more ethical world.

“News is the process of shaping collective attention, bringing to light what is important when most other things remain in the darkness.” — Matt Carlson

Robot journalism is an opportunity to make journalism more human because it contributes to strengthening the competitive intelligence of newspapers. It diversifies the craft and allows new content to be produced fast, with a great accuracy attracting new models of consumption and distribution. I believe AI will play a major role in the creation of news in the future but we’ll have to be careful and prudent about its use and development rather than indifferent and dismissive.

Like Karl Marx once stated: “machines built through technical advancement will displace humans in the workplace”. I wonder then how far are we from reading at the bottom of every news item a disclaimer that states: “This story was written by a robot.

And even though the technology is not very sophisticated and still has a lot of limitations, we probably are not that far away from a time before machine-written software takes up the central role in the news landscape and becomes part of our daily news routine. Fortunately, for now, it’s only smelly, hairy Homo sapiens jostling elbows to get the few gigs left in the business.


This article has been written by a human.

This article is the second part of a series on issues about the future. Read below the third part: