What’s to come for journalism and artificial intelligence? GNI and Polis report

2020 will be a big year for AI in journalism. How have publishers evolved and what do they see ahead?

REUTERS/Reuters TV

By Sahar Amer | Feb 20, 2020

Amid rising fears that artificial intelligence (AI) will threaten journalists’ jobs and take over the newsroom, the Journalism AI report – a project by Polis in collaboration with Google News Initiative – sought to find out how exactly AI technologies are being applied to journalism. The 71 news organizations from 32 different countries in Europe, North and South America, Asia and Africa that took part in the report, mentioned that the biggest AI adoption challenges are:

Financial resources (27%)

Knowledge and skills (24%)

Cultural resistance: reluctance to change work habits and fear of losing job (24%)

Lack of strategic managerial insight (17%)

However, AI is a ‘significant part of journalism already but it is unevenly distributed’ and news organizations are already applying aspects of intelligent technology in their operations, to help them work more efficiently and improve monetization.

What’s currently happening?

“One of the key aspects of AI and journalism is that it allows the whole journalism model to become more holistic, with a feedback loop between the different parts of the production and dissemination process” — Charlie Beckett, founding director, Polis

The report identified three main areas where newsrooms are using AI:

News gathering
News production
News distribution

Artificial intelligence systems can be useful in helping newsrooms to categorize content or information at scale for different news gathering purposes. For example, since 2015 The Associated Press have been using a management tool, SAM, which algorithmically sifts through social media platforms to alert the newsroom on likely breaking news events. Alongside making the news gathering process easier, AI is also used to spot stories that are previously undocumented and would otherwise go unnoticed. El Universal teamed up with Google News Initiative to use machine learning in order to quantify and visualise the areas across Mexico where journalists were not reporting on violence or threats in fear of their own and others’ safety.

A map of unreported murders across Mexico that were identified through El Universal’s project. Source: GNI

On the news production side, AI systems are assisting with a range of cognitive labour, such as: grammatical and formatting errors, fact-checking and verification, and scheduling posts to send out through social media. The Finnish public service broadcaster, Yle, have developed a personal news assistant bot, Viotto, which displays news recommendation pop-ups through the lock-screen on Android phones and produces text and illustrated content for the news app, Twitter account and newsletters.

“AI is used for routine, commodity journalism but it can also be combined with human journalists to tailor mass data sets to specific audiences.” — Charlie Beckett, Founding Director, Polis

Many news publishers are already using AI systems alongside journalists to produce general data stories. One of the most recent is The Guardian’s ReporterMate, an open source automated system that, when given a data set and a story template, creates formulaic news stories – like this one about Australian political donations. Using an AI software that can generate data stories can enable journalists to spend time on more deep and meaningful reporting.

Access and strategize

“AI is innately networked. It requires a skill set, knowledge base, defined roles, systematic appraisal, and a culture that relates the technologies to editorial or marketing.” — Charlie Beckett, Founding Director, Polis

Journalism AI report

The report concluded that responses from participants signalled a distinct lack of strategic planning around AI. When it comes to implementation, deciding what approach a news organization should take will differ according to the nature of the organization and what stage they are at. However, two factors are always decisive – time and money. Outsourcing various partners of specialists may be time consuming and cost a lot of money, but building an in-house team will pay off in the long run. According to Charlie Beckett, Founding Director of Polis, in order for AI to be implemented successfully within the newsroom there needs to be a foundation of good knowledge, sufficient skill set, and defined roles. For example, The Washington Post (Heliograf), have homegrown all their AI and have employed a team of 250 tech heads. Emilio Garcia-Ruiz, Managing editor of WP, says “The big advantage of having this in-house team is that you can work together with journalists who can provide the correct journalistic context.”

The question of ethics

“The majority of respondents had confidence that overall, the impact would be beneficial if news organisations retained their ethical and editorial stance.” — Charlie Beckett, Founding Director, Polis

Although there is huge transformational potential when it comes to AI, there’s also wide discussion around its ethical issues. It’s therefore important for news organizations to consider ethical factors when using AI technology in order to keep journalist accountability and integrity intact.

Journalism AI report

One fifth of the participants from the report were less worried about the implications of AI and journalism, and were more excited by the prospects.

“I am more excited than concerned regarding the impact of AI. As opposed to the current negative sentiment towards AI, these technologies will augment the newsrooms and save valuable resources to be directed toward serious issues that require the attention of journalists.” — Anonymous, The Journalism AI report

However, debates surrounding the maintenance of journalistic codes of ethics are still necessary. The context in which AI is being utilized must influence the modes in which artificial intelligence tools are utilized. With dwindling public trust and confidence in the media, it’s important that publications do not dent an already tarnished reputation. Six key areas were identified through discussions with the participants on their current thinking about AI, ethics and journalism:

Economics: savings or investment?

Algorithmic bias

Misinformation and ‘filter bubbles’

Enhancement of editorial decisions and transparency

Balancing artificial and human intelligence

The role of the technology companies

Where AI is used for efficiency, it could also possibly devalue human labour which could have implications for ‘quality’ journalism and the value it has to the public. Algorithm biases could also have huge implications for credible journalism by enhancing production bias of a news organization and inequality around profiling certain gender or racial groups.

What’s the future for AI and journalism?

“The future development of AI in general is not a smooth upward curving graph. There are still fundamental debates within the AI world about the best pathways forward.” — Charlie Beckett, Founding Director, Polis

As AI continues to permeate through different areas of the newsroom, respondents from the survey mentioned that the challenges they anticipate in the future focused on training, education, and literacy in the newsroom (46%) and the need for recruiting people with new skills (43%). As it’s still in early stages of integration, newsrooms will keep benefitting from a tried and tested approach to AI aiming to run more efficiently and sustainably with ethics at its core. Sign up to Reuters Community, for exclusive insights that keep you innovating.