Artificial Intelligence enters newsrooms; are jobs at stake?

New Delhi: The first wave of Artificial Intelligence (AI) has started creating ripples in the newsroom – in chatbots, automated reporting systems and machine learning techniques which are sifting through massive data sets to write initial news reports.

For those who still believe that AI and automation will not hit jobs in the media industry, just consider the kind of investments that media houses are now making in AI for the newsroom. Clearly, the next five to ten years would be decisive.

Google has provided British news agency Press Association $805,000 to build software that will gather, automate and write nearly 30,000 local stories a month. Dubbed as "Radar" (Reporters And Data And Robots), the software will "automate local reporting with large public databases from government agencies or local law enforcement."

South Korea's Yonhap news agency has introduced an automated reporting system "Soccerbot" to produce news on football games. Associated Press and Thomson Reuters are using machine learning algorithms to write stories and The New York Times plans to automate its comment moderation.

So whose jobs are at stake as AI-powered software becomes common in the newsroom? "For simple, factual-based news reporting and gathering – it is possible for AI-powered bots to do that. However, it's less likely for point-of-view type of stories or feature articles that require more human input," said Xiaofeng Wang, Senior Analyst at global research firm Forrester.

Personalised communication at scale is one of the strengths that bots can deliver on more efficiently. "There are already bots designed for research purposes, but they act more like assistants to humans. When designing these bots, humans can define the purpose clearly and imagine a personality for it but the AI technology today still has limitations," Wang told IANS.

Carl Benedikt Frey and Michael Osborne, researchers at the University of Oxford, have predicted that the combination of robotics, automation, AI and machine learning would impact journalists too, among others. According to a survey by the University of Oxford and Yale University, AI would outperform humans in many activities in the next 10 years, such as translating languages (by 2024) driving a truck (by 2027), writing a bestselling book (by 2049), and working as a surgeon (by 2053).

"There are dynamic changes happening and we as journalists can see that. AI and robots will write stories in the future and they will get better at doing it as humans will fine tune it," noted Ramesh Menon, a senior journalist and author. "But the best stories will come from writer-journalists who can put in fine detail, empathy, drama, colour and analysis into their stories. What is good is that we can tomorrow get robots to do the routine stuff that today takes 80 per cent of the journalists' time," Menon told IANS.

This can help reporters and editors concentrate on big ticket stories that require a lot of footwork in terms of getting to the right people, getting them to talk, analysing the present and even talk of the way forward, he emphasised. Researchers claim that there is a 50 per cent chance that machines will outperform humans in all tasks within 45 years. Yonhap's "Soccerbot" would write stories related to the English Premier League (EPL) football tournament.

The system had a successful test in covering all games of the 2016-17 EPL season. It produced a total of 380 automatic experimental articles, each within one or two seconds of a game's end. The "Radar" software project at Press Association, scheduled to be launched in early 2018, will hire five journalists who will help identify datasets and curate and edit news articles generated by "Radar".

According to Rajeev Vaid, Chief Operating Officer and Co-Founder, machine Intelligence platform Kognetics, AI is now being used by leading global media houses for automation of press coverage. "In future, we expect to see a wide-scale use of automated systems to analyse content quickly, in real time, and produce insights in a consumable and usable format to the journalists," Vaid told IANS.

The initial impact of AI on journalism would be seen at four levels: Content gathering, content creation, moderation and scale. "AI can be used to scan through structured and unstructured news corpus to gather information restricted to an area, territory or region. Similarly, it can be used to disseminate content to a targeted section with local news via Internet, social and print media," Vaid stressed.

For Atul Rai, CEO and Co-Founder, Staqu, a Gurgaon-based AI startup, AI could be used in most of the domains which are data excessive but identifying fake news is another big challenge today for humans as well as bots. "Media is a domain where plenty of data is collected and created and, therefore, it is easier to use these data points to create stories. I would reiterate that the problem set is not to create news or content but is filtering out the malicious, fake and rumoured content or news," Rai told IANS.

As the media fraternity prepares for a new reality, the news business would only ramp up incorporating the AI-based tools into the newsroom. "Though many of us are apprehensive today, it may not be such a bad thing. Good journalists will always have jobs; no one can replace them in the world of tomorrow as they would be needed more than ever," Menon said. For Wang, AI would help journalists work more efficiently if they know how to use (not create) these bots. "Bots are not ready to replace journalists yet. Replacing the journalist's assistant? Maybe," she added.