Speakers :
- Lisa Gibbs, automation / Al newsroom Lead at Associated Press / responsable de l’automatisation et de l’intelligence artificielle à Associated Press.
- Alexandre Elkrief, Co-Founder and President of Knowhere News / co-fondateur et président de Knowhere News.
- Moderator : Frédéric Filloux, Associate Professor at Sciences Po Journalism School / professeur associé à l’Ecole de Journalisme de Sciences Po.
Why AI can be useful for journalists
Technologies have seriously changed the way the media work: for instance, we need to produce multiple stories that match the multiplicity of different platforms we have at our disposal. In this context, the way we choose to use artificial intelligence in our favour is an exciting topic. We create algorithms for YouTube, Snapchat…and any other social network, really. Not only do we use it to recognize news, but also to filter it, alert the editors of the existence of certain news, and so on. « We are in the era of insight detection”, Lisa Gibbs sums it up.
AI is also very useful to help journalists with video transcription, saving us hours of work.
Why AI can be seen as a threat for our profession
Knowhere News uses AI to produce news content: “Our technology scours the internet, discovering stories and identifying narratives, evaluating the factual claims and bias in reporting. This analysis informs a set of algorithms that write our three perspectives on every controversial story », their website explains.
But moderator Frédéric Filloux is skeptical about the approach: “Don’t you think you are oversimplifying journalism? Don’t you think journalism is way more complicated than that? Don’t you think you are adding bias where our job is to get rid of it? », he asked. “People expect from journalists to present the news as neutral as they can. AP will never do what Knowhere news does, which is aggregating and recreating stories thanks to multiple sources” added Lisa Gibbs.
“How can AI help us get a more accurate diagnosis of the current mood or state of a population, for instance during elections or protests?” is another question posed by Frédéric Filloux. The point is, he argues, that AI clearly lacks important human qualities when faced with these situations.
Even the public was doubting this new practice: “How can you trust what it is written when it is based on invented stories? », asked someone in the room. Alexandre Elkrief from Knowhere News finally recognized that what his platform’s offer was something biased. While artificial Intelligence, the speakers and the public summed up, does offer some exciting possibilities for the journalistic practice, it can not yet be considered to be an extremely trustworthy tool for journalists.
Crédits photo : Jules Boudier