Scientific Foresight (STOA) By / June 25, 2019

Artificial intelligence, machine learning & automation: what future for journalism?

Some 70 enthusiastic young journalists from all over the EU met in Strasbourg from 4 to 7 June 2019, for the first edition of the European Youth Science and Media Days (#eysmd2019) – the summer school on ‘Artificial intelligence and journalism’ organised by the European Science-Media Hub (ESMH/STOA) in cooperation with the European Youth Press network of media-makers (EYP).

Written by Vitalba Crivello,

first edition of the European Youth Science and Media Days (#eysmd2019)Some 70 enthusiastic young journalists from all over the EU met in Strasbourg from 4 to 7 June 2019, for the first edition of the European Youth Science and Media Days (#eysmd2019) – the summer school on ‘Artificial intelligence and journalism’ organised by the European Science-Media Hub (ESMH/STOA) in cooperation with the European Youth Press network of media-makers (EYP).

The event was the first of its kind for Parliament’s Science and Technology Options Assessment (STOA) Panel and the European Parliamentary Research Service (EPRS), and it proved a success in terms of participant engagement and feedback received so far. The idea behind this format is to turn it into a regular forum, offering young media-makers the opportunity to learn about the latest technology tools and practice using them for their work. The results of the satisfaction survey distributed to the participants will help the ESMH to collect suggestions and remarks to take on board for the next edition.

Ten artificial intelligence (AI) experts (researchers, policy-makers, journalists and media representatives) shared their experience and reflections on key aspects of the intersection between AI and journalism:

  • What does the rise of automation mean for media-makers?
  • How does AI affect journalism?
  • What role can algorithms play in this changing context and what are the possible ethical implications?
  • How will AI and humans combine in the journalism of the future?

 The programme was dense and diverse, ranging from thematic panels, hands-on training and case study presentations (including an AI tool demo), workshops for participants and a virtual reality experience.

Thematic panels focused on ‘AI, EU & ethics’, ‘AI in the newsrooms’ and ‘AI & algorithm literacy’. The European approach to AI aims at putting Europe in the lead globally by deploying only ethically embedded AI, while promoting innovation and investment. Robots can definitely play a game-changing role, as they accelerate the interpretation of information, challenging the professional practice of journalists and researchers. However, the fundamentals of providing sense and critical assessment with ethics, integrity and sound judgement remain valid and relevant. AI in newsrooms is real, but AI can be made ‘uncool’ again, according to Mattia Peretti (LSE, JournalismAI), through understanding of the challenges and opportunities presented by the adoption of AI-powered technologies in newsrooms and what AI can, should and should not do for journalism.

Other issues addressed by the thematic panels and the accompanying discussions with the audience included:

  • Automation and AI have already made their way into newsrooms, helping journalists do their work quicker or better. State-of-the-art algorithms can already provide decent, brief and readable summaries of research papers, opening up the possibility of automated science press releases and news.
  • Notwithstanding the fact that growth and development of AI is spreading into creative domains, the emerging challenges relate mostly to the distribution of content using AI, rather than to the use of AI in content creation.
  • Editorial thinking helps to hold algorithms accountable for possible consequences. An algorithm is simply a set of instructions, similar to a recipe. Like any set of instructions, they reflect the biases of the people who create them. The tech industry might not have a codified body of ethics, but media professionals do, and they are committed to truth, transparency, accountability and harm minimisation.

In algorithmic literacy, the key question is: what can or should be automated and what are inherently human tasks? ‘Augmented journalism’ is a novel concept, which differs from the idea of AI and machines that are smarter than humans.

The ‘machine learning’ hands-on training offered journalists the opportunity to learn how machine learning really works and also how to distinguish facts from myths, to cut through the hype about AI, as well as the misinformation about what it can and cannot do.

Participants also experienced a demonstration of INJECT, a new digital tool (EU-funded) available to newsrooms to support journalists in writing more original pieces, by using different technologies to enhance and empower journalists faced with less time, and fewer resources.

A full session was devoted to case studies from the European Institute of Innovation & Technology (EIT) community, showing that EIT is running one of the biggest AI programmes in Europe with education, innovation and entrepreneurship activities, helping to shape the future of Europe’.

Finally, the virtual reality (VR) experience included a VR cinema screening a selection of three films on AI for participants (‘Alteration’, ‘Merger’, and ‘I saw the future’). Moreover, participants could use three ‘room-scale’ VR stations to test a space-themed experience (Mercury project), a vertigo-themed experience ‘spheres’, and an artistic experience to paint in 3D.

The afternoons were devoted to nine working groups where the young journalists produced outcomes on selected topics discussed during the summer school. These groups produced several interesting media products on AI, which they showcased on 7 June. Stay tuned to the ESMH webpage to see them!

For more details on the event:


This slideshow requires JavaScript.


Related Articles

Discover more from Epthinktank

Subscribe now to keep reading and get access to the full archive.

Continue reading