In a time where reporting about COVID-19 is no longer merely a prerogative of science journalists, the importance of empowering and training journalists, including those without specialist scientific knowledge is more relevant than ever.
Among the several outputs to be released by the QUEST Project in the second half of 2020 is JECT.AI, a tool for journalists based on AI support. This tool has been created to allow journalists to report about science more effectively.
The development of JECT.AI is led by QUEST partner City, University of London. JECT.AI takes advantage of a variety of automation systems in order to support journalists incorporating writing practices generally associated with good science journalism.
Three of these practices include:
(I) Using more diverse scientific information sources to develop stories, such as science alerts, scientists’ tweets, science newsletters, and general news sources;
(II) Tailoring scientific content to reach and engage specific target audiences;
(III) Effectively explaining science to these audiences.
This AI-powered tool can research and present information from thousands of diverse sources, including peer-reviewed papers, views expressed by science experts and journalists, and stories published in newspapers. JECT.AI discovers this information and presents them to a journalist in a common format in one place.
Another feature included is JECT.AI's audience “personas”. Namely, it presents a range of fictionalised characters whose personalities embody the different typical engagement profiles. This list of personas is meant to help journalists target their articles, matching the contents to the different audiences’ interests and backgrounds.
Moreover, the JECT.AI system can provide a set of metaphors, automated through predefined rhetorical relations, such as context, example, motivation, and cause. Said metaphors, e.g. greenhouse effect and herd community, are to provide guidance to journalists and allow them to better communicate complex scientific subjects with the lay public.
Some of JECT.AI’s features are already available at this link.
[Please note: JECT.AI is the new name for the tool, previously known as INQUEST]
This project has received funding from the European Union’s Horizon 2020
research and innovation programme under grant agreement No 824634