How Le Monde is taking on fake news

Fake news has become an online pestilence, and Le Monde wants to bust it wide open. That’s why the French newspaper’s 13-person fact-checking unit Les Décodeurs has been quietly developing a way to automate how to root out false news, at scale.

Samuel Laurent, who heads up the Le Décodeurs division, is no stranger to false news, having already debunked several major hoaxes surrounding terrorist attacks in France this year. And yet, with numerous news sites and social media pages actively spreading hoaxes in France, Laurent believes that the situation is becoming “toxic.”

Although fact checking is a core part of all quality newsrooms, the sheer volume of sources from which information is spread, makes verification a strain on internal resources. Added to that, publishers with global teams can have four reporters in different countries overlapping on fact checking, which is a time drain. Automation would help speed up the process, and staunch how quickly wrong facts can be parroted by other online outlets and content farms. At least, that’s the theory.

“Some level of automation is so important for time saving, and the amount of resources we need to fact check. Sometimes you feel you are trying to empty the sea with a glass of water,” said Laurent.

The plan is to build a hoax-busting database, which incorporates information on which sites are fake and which are verified, trusted sources, and readers can access via Google and Firefox Chrome extensions. The idea is that once a user has downloaded the extension, when they come across articles online a red flag will appear if the site or news is deemed fake, yellow if the source is unreliable or green if it’s ok.

Laurent’s team has also been working with an external set of data scientists to explore how to automate spotting hoax news, and the hope is that the end product will resemble a search engine that can query relevant databases to spot and provide context around false news – a project that’s received funding from Google’s Digital News Initiative.

“If we have a politician talking about unemployment rates in Britain for example, our tool could understand we are talking about unemployment in that country and will go themselves and produce data around that politician’s claim. So we are currently working on plugging the tool to sources like EuroStats and other national statistics tools,” said Laurent.

The first phase of the tool will launch in January 2017, and Le Monde will run marketing messages across its sites, guiding people to download the extension and signposting articles and tutorials that give information on how to check stories and ensure what they’re reading is reliable. Laurent hopes to have the complete hoax-busting platform, up and running by the end of next year.

Laurent has plans to expand the hoax-busting database beyond France’s borders. “Our goal is to be open source, so that everyone can use it. And we hope to have a bigger database by sharing our database of news sites and other fake news from other countries,” he added.

Educating younger generations around how to spot false news, and distinguish between sites peddling nonsense, is going to be hugely important, added Laurent. That’s why Le Monde is also actively seeking partnerships with the education ministry in France and with colleges and schools, to help set up resources and tools to educate younger generations on how to spot hoax news from real news. “That’s something that’s already ongoing in France, but not as much as it should be,” he added.

https://digiday.com/?p=211686

More in Media

Digiday+ Research: Publishers expected Google to keep cookies, but they’re moving on anyway

Publishers saw this change of heart coming. But it’s not changing their own plans to move away from tracking consumers using third-party cookies.

Incoming teen social media ban in Australia puts focus on creator impact and targeting practices

The restriction goes into effect in 2025, but some see it as potentially setting a precedent for similar legislation in other countries.

AI Briefing: Amazon’s new Nova models boost AI model efficiency, accuracy and variety across AWS

One of the most buzzy debuts was Nova, a suite of six new AI models that include understanding and creating text, images and videos.