Skip to main content

How AI enables reporters, photo-journalists and broadcasters to humanise content

AI
(Image credit: Shutterstock / Peshkova)

AI has played a role in breaking some of the biggest international news stories of recent years. A key example is the ‘Panama Papers’ exposé, where machine learning helped an international team of researchers to identify loan agreements in more than 13 million records that were leaked to the press. Journalists were able to ‘follow the money’, exposing the practices of offshore tax havens and the businesses taking advantage of tax loopholes.

Machine learning also played a valuable role in the Implant Files investigation. Here, it sifted through reports sent to the US Food and Drug Administration (FDA), and helped to uncover patient deaths potentially caused by faulty medical devices.

For now, most machine learning is supervised by software engineers. But in the future, ‘domain experts’, including journalists, will play a greater role. If machine learning can help doctors assess valuable information from x-rays and CT scans in order to triage patients, then it should be able to further assist journalists. As the opportunities for big data investigations increase, reporters may need to acquire skills to instruct machine learning research.

Categorizing content will transform photo agencies

Digital photography has transformed the business models of photo agencies. Gone are the days of thousand-dollar invoices for the exclusive rights to a photograph. Instead, agencies are moving to a high volume, low margin, royalty-free model.

For this to be viable, many agencies are adopting computer vision technology to tag images, a form of deep learning that can tag images with their contents including objects, people and emotions. This makes it easier for broadcast media to find, and then purchase, images to illustrate their own content.

Such technology includes facial recognition, essential when classifying archives that contain millions of photographs gathered over dozens of years. The most advanced systems also include an ‘aesthetic ranking feature’, which enables photographers to filter their best images based on composition, depth of field, position of subject, contrast and other factors.

Machine learning, in this case deep learning, enables specialists to focus on their core strengths. Photo-editors have more time to examine the narrative possibilities of new pictures. Photographers, who may take thousands of images at a single event, no longer have to sift through every single image. Instead the software recommends a much smaller set of images to work from.

Other organizations who can benefit from this approach are news agencies, such as Netherlands–based ANP, which uses artificial intelligence and facial recognition to tag photos faster and more accurately. Client searches now return more relevant images, increasing revenues while reducing operational costs.

Editors will spend far less time on data entry and more time making critical editorial decisions. “Our journalists can just be journalists again,” says Patrick Rasenberg, product manager photo, ANP. “It’s a good example of where a new technology makes the workplace more human and a rewarding place to work.”

Using automation to build stories

AI can write news articles - just not very complicated ones. In most cases, such systems rely on what is known as robo-journalism, the automated writing of stories based on structured data. This works well for deadline-driven stories based on sports results, financial news, weather and elections, to name but four.

The Radar News Service in the UK is a good example of this approach at scale. Launched in 2018, its five reporters filed 250,000 articles in the first 18 months of service. The journalists use specialist NLG (natural language generation) technology to draft articles based on data sets released by the UK government.

Using their investigative skills, the journalists identify data sets from which they can derive a story and then build a template into which the data and standard phrases can be assembled. Stories are then published to subscribers, especially local news outlets, who may publish the original content or use it as the basis for their own reporting.

For journalists to flourish in this environment, being able to work with automation templates is essential. Some simple programming knowledge is also required. It is another good example of AI empowering news reporters, helping them to be more effective at their jobs.

Building trust in the age of social media and deep fakes

Technology has impacted fact checking in several ways. It supports the collaboration of literally hundreds of journalists and fact checkers across different offices. In the case of the Implant Files story, fact-checking involved a team of 11 people manually sifting the results, to make sure that every case flagged by the algorithm was correctly identified.

Machine learning has a role to play in other fact checking scenarios, either supporting human specialists, or validating information itself. On the one hand, it can help separate out sentences with claims, making it easier for fact checkers to focus on statements that require validation. On the other, it can verify claims autonomously, checking against databases of information in real time.

Machine learning also plays a valuable part in the technological arms race against fake news and its propagandists. As the volume of misinformation increases it will play an ever-greater role detecting such content and preventing it from contaminating articles authored by journalists and robo-journalists.

“Our new mission statement is ‘Start With the Facts’,” says Patrick Rasenberg from news agency ANP. “Our clients place absolute faith in the integrity of our image archive and the accuracy of the new software will ensure that we preserve and build on that trust.”

As the above examples show, there are two main areas where artificial intelligence has a role to play in today’s newsrooms: helping to deliver more relevant content to users, and improving business efficiency by undertaking time-consuming research and fact checking activities.

Put it another way, AI will not replace journalists, it will empower them. At a time when free speech and honest broadcasting are under threat, it frees up reporters to focus on what they do best: nurturing sources, coordinating research, and assembling articles using professional expertise and creative skills. Computer vision solution will take over time-consuming processes so journalists can focus on what matters the most, the human side of reporting.

Dr Appu Shaji, founder, Mobius Labs