Can innovation save the truth?
Features
How Innovative AI Tools Are Tackling Misinformation Head-On
Last week saw the much-anticipated Harris-Trump presidential debate take place, hosted live on the ABC news network. During the 90-minute broadcast, the two candidates answered a series of questions neither had seen beforehand but could anticipate. This meant that, for their teams, there was ample opportunity to get their facts straight.
One of the key issues raised by the debate was the dangers of misinformation, despite it not featuring specifically in the list of questions asked. In his already infamous response to Kamala Harris’ claim that people leave his rallies early, Donald Trump repeated a debunked claim that Haitian immigrants in Springfield, Ohio, have been stealing and eating people’s dogs and cats. The claim had previously been given oxygen on social media, after a video of a Springfield resident accusing immigrants of stealing ducks from a local park for food went viral, followed by a series of AI-generated memes of Donald Trump saving cats and dogs.
The accusation was fact-checked in real time during the debate, but Trump refused to concede the point. Cue another avalanche of social memes and reels.
News organisations have a number of credible tools and organisations to draw on for fact-checking, but it remains up to the individual user of a platform or reader to draw their own conclusions about content that does not come from a credible source. In August, Meta discontinued the use of CrowdTangle, a social media monitoring tool that was used to track misinformation on Facebook and Instagram and replaced it with a new tool called Meta Content Library (MTL), which some commentators are saying falls short of its predecessor. Elon Musk has championed community-based fact-checking on X, which can be effective in some cases, for example countering false claims with scientific sources, but not where there is an echo chamber of inflammatory sentiment.
So what can be done?
In June, we featured a story about a US not-for-profit, Meedan, which runs programmes that build capacity and digital literacy for news organisations and their audiences, primarily focused on major events such as elections, climate and public health emergencies, and armed conflicts. It also provides organisations with open-source software to enable efficient fact-checking.
It also undertakes valuable research into how AI can be leveraged in the fight against misinformation. In November last year, The Rockefeller Foundation awarded The Solutions Journalism Network (SJN) a grant to work with Meedan to develop a large language model based tool to evaluate and classify news stories. Initially the tool will be used to identify climate solutions stories the goal is to roll it out for any issue. In terms of climate reporting, SJN hopes to counter the apocalyptic tone of much of the reporting related to climate and to focus on solutions instead.
“The AI classifier that Meedan is developing will make it possible for people to search specifically for news stories that show potential to respond to real-time challenges related to climate change,” says SJN CEO David Bornstein. “By making it possible to systematically identify rigorous reporting about climate adaptation, mitigation and resilience efforts from far and wide, this tool promises to speed up the global learning needed to make headway against the climate crisis.”
What do you think?
It’s a positive step but more is needed to counter the onslaught of misleading content online. We want your views – do you fact-check what you read online? If you want to discover real solutions that are changing the world for the better, have a browse of the Springwise Library.
Words: Angela Everitt
16th September 2024