Media futurist Hossein Derakhshan predicts that AI-generated writing and similar tools spell extinction for most newsroom jobs.
“AI technologies are now capable of doing almost every aspect of the news practice with a fraction of the costs in the long term,” Derakhshan observes.
“The AI capture of newsrooms will kill more jobs than we can imagine,” Derakhshan adds. “It will also kill many newsrooms who can’t afford to invest in AI technologies to lower their costs in order to survive.”
Derakhshan’s silver lining: Journalism is reinventing itself in a form that can complement AI-generated writing — one he calls ‘post-news journalism.’
Essentially, it’s the kind of long-form, affective narrative that is the hallmark of journalists like Bob Woodward, Errol Morris and Michael Barbaro, Derakhshan says.
Derakhshan is a London-based media researcher and a former fellow at Harvard University’s Shorenstein Center on Media, Politics and Public Policy.
His view differs markedly from fellow academic Nick Diakopoulos. He sees AI journalism as changing the nature of newsroom jobs – but not eliminating them.
“Human work will be hybridized – blended together with algorithms – to suit AI’s capabilities and accommodate its limitations,” Diakopoulos observes.
Diakopoulos is assistant professor, communication studies and computer science, at Northwestern University.
In other AI-generated writing news:
*Expect More Press Releases to Be Written By AI: PR agencies will increasingly turn to AI to write press releases in coming years, according to Katie Paine.
She’s CEO of Paine Publishing, which specializes in analyzing the impact of its clients’ marketing and press relations.
“AI and data analysis will be an ever-larger part of the communications toolkit — not just for measurement, but also for responding to crises, managing risk and even writing press releases,” Paine observes.
Paine is one of 20 PR industry experts PR Daily interviewed for prognostications on what the PR industry can expect in 2020.
*Students Worldwide Swarm Course in Data Journalism: Nearly 13,000 students from across the globe signed-up for a course in data journalism offered by the Knight Center for Journalism in the Americas.
“We were stunned at the enthusiasm for the course,” says Simon Rogers, one of the course instructors and data editor on Google’s News Lab Team.
The six-week stint took students from 160 countries through “the entire pipeline of producing a data story,” Rogers says.
One of the most widely known applications of data journalism is the AI-generation of vivid graphs and other illustrations from info databases.
But data journalism also refers to the AI-generation of articles from databases.
*Another Fan of AI-Generated Writing in Marketing: Email marketing expert Doug Morneau has fully embraced the use of AI, Big Data and machine learning.
The result: Morneau has nothing but praise for the technologies.
“I’m already using all three for email marketing, email subject lines, Facebook advertising and copywriting,” Morneau observes. “The ability to leverage machines to create and test hundreds of instances of advertising, images, and offers will now be at your fingertips.”
Morneau is one of 20 PR industry experts PR Daily interviewed for predictions on what the industry can expect in 2020.
*BBC Covers Election With AI-Generated Writing: One of the most highly respected media outlets in the world used robot software to tell the story of Britain’s 2019 election results.
Robert McKenzie, editor of BBC News Labs, says AI-generated writing was used to report election results for 650 individual constituencies in the U.K.
Each story was generated using a story template, which was then populated by data generated by a specific constituency’s election results.
“The machine selects particular phrases or particular words in response to precise pieces of data,” McKenzie says. “So you can write everything if you want to in ‘house style.'”
*EBU Releases List of Key AI Tools for Reporters and Writers: The European Broadcasting Union has released a representative sampling of AI journalism tools content creators can use to help automate writing and related tasks.
Among the most interesting is ReporterMate, an AI-generated writing program. It’s open-source software that’s free to download and use.
Other notable tools include Newswhip, which detects trends in social media, Dataminr, which looks for breaking trends on the Web and Botslayer, which exposes social media personas that are fronts for robots.
Still other tools include Truepic, which detects fake photos and videos and Graphic, which auto-creates visualizations from data.
The list is part of a comprehensive November 2019 report on how AI-generated writing and similar tools are remaking journalism, released by the European Broadcasting Union.
*News Outlets Collaborating with Universities on AI-Generated Writing: Big media houses are partnering with universities to bring AI tools to the newsroom, according to a new report from Charlie Beckett.
He’s director of the Media Policy Project, sponsored by the London School of Economics and Political Science.
Specifically, Beckett says news outlets are looking to universities for help in:
*Researching and developing AI tools
*Recruiting AI experts from the university community
*Expertise in developing and implementing specific AI news projects for the newsroom
One high profile example: The Washington Post, which is working with Northwestern University to help AI automate coverage of the 2020 U.S. election, according to Beckett.
“It is a model for how newsrooms and universities can find ways to work together for mutual benefit,” Beckett observes.
Beckett examines AI collaborations between news media and universities as part of his November 2019 report, “New Powers, New Responsibilities: A global survey of journalism and artificial intelligence.”
*A Window on AI and Investigative Reporting: The Columbia Journalism Review offers an intriguing Q&A with John Keefe.
Keefe plays an integral role in the Quartz AI Studio, established about a year ago to help journalists understand when and how to use AI for investigative reporting.
Quartz AI Studio is funded by the Knight Foundation.
“We’re super-interested in helping individual reporters as they solve individual problems,” Keefe says.
The Keefe Q&A is part of an in-depth look at the overall impact of AI on journalism by the Columbia Journalism Review.
The series features Q&A’s with:
*Marc Lavallee, The New York Times
*Nathalie Marechal, Ranking Digital Rights
*Kaveh Waddell, Axios
*Daphne Keller, Stanford University
*Will Oremus, One Zero
*Tiffany C. Li, Yale University
*Aviv Ovadya, Thoughtful Technology Project
*Nick Diakopoulos, Northwestern University
*Charlie Beckett, London School of Economics
*Karen Hao, MIT Technology Review
All told, it’s a great set of Q&A’s for a quick study on how AI is impacting journalism.
*Data Journalism Bolstering Credibility of South Korean Media: South Korean journalists say their use of data journalism is helping them regain the trust of the general public.
Specifically, data-based journalism enables them to release the raw data associated with a story directly to the public. That imbues their stories with greater transparency, they say.
“Every reporter needs some level of data literacy and analytical know-how,” says Yun Won Choi, a data journalist at Newstapa, a South Korean news outlet. “That’s what will help them write articles that provide unique insights.”
Data journalism is an umbrella term for computerized visualizations and news stories, which are generated from the manipulation of databases.
Data journalism is best known for visualizations – graphs, charts, infographics – generated from databases with the help of AI.
But the form also includes AI-generated news stories wrested from data.
News service Radar, for example, regularly produces thousands of news stories each month drawn from British government databases.
Currently, the UK-based service is squired by six journalists, who create AI story templates, which reach down into British government databases to unearth localized news stories.
Using the method, Radar was able to hyper-localize a government database chronicling crime across Great Britain and come up with a story entitled, “Hackney: The home of bike theft in London.”
Hyper-localization and hyper-personalization are some of the great strengths of data journalism: The ability to use AI to quickly explore a vast database, and come-up with a news story or visualization that applies to a highly specific location or a highly specific term.
Another example: Radar was able to reach down into a giant database on National Health Service spending and come up with this hyper-personalized story: “Figures reveal NHS spends millions prescribing gluten-free food for patients.”
*AI-Generated Holiday Cards: Not Ready for Prime Time: Business consulting firm R/GA London fed AI-generated writing software hundreds of holiday songs, greetings and sayings so it could auto-produce sayings for greeting cards.
The result: Some less-than-spirit-invigorating sentiments, including “Don’t Get the Tinsel,” “Let Earth Be Freezin’,” and “Stir the Eggs, Maria.”
With all due respect to Maria’s eggs, the experiment is not a reflection of AI-generated writing at large.
Instead, it’s simply one result, created by one team of programmers, using one form of AI-generated writing software and one type of database.
Put another way: R/GA London’s AI experiment yielded dreck. But another experiment could have yielded entirely different results.
Feel free to send a link to RobotWritersAI.com to a friend or colleague.
*Also on RobotWritersAI.com — Evergreen Article:
*AI-Created Newsletters: On The Cheap
–Joe Dysart is editor of RobotWritersAI.com and a tech journalist with 20+ years experience. His work has appeared in 150+ publications, including The New York Times and the Financial Times of London.