Can AI-Generated News Help Reverse the Declining Level of Trust in News?

AI  Communication and Technology 

With the generative AI boom, comes a consideration of trust in content created by AI. Gallup’s  annual media trust poll revealed that more than a third of Americans (36%) have no trust at all in mass media, a daunting audience to try to infuse new means of content in.

As more news outlets turn toward integrating AI content in their platforms and distribution models, researchers are curious if AI-generated news can weave a new fabric of news consumers and improve trust in community organizations in the process.

A team at the University of Florida College of Journalism and Communications (UFCJC) led by Seungahn Nah, Dianne Snedaker Chair in Media Trust and research director for the Consortium on Trust in Media and Technology, has pondered whether credibility and trust in news are sacrificed when articles are AI-generated and if public nonprofit perception could benefit from exposure to that content.

The Consortium assembled the AI Research (AIR) Group – a collaboration between UFCJC and scholars from the University of South Florida and the University of California, Los Angeles – which is conducting studies using a completely AI-generated news site, built in-house by the group, called MyMiamiNews.org. The website focuses on nonprofit and voluntary organizations in Miami and its neighboring communities.

Complete with full-text articles and AI-generated art for the stories, the site is presented as a community-centered, nonprofit news platform uplifting community organizations and serving informational needs.

The group tracked user login data and conducted weekly surveys on user interaction with the AI-generated news.  They found that the more exposure audiences had to the AI-generated news site and the more times they used it, the more likely they were to perceive AI-generated news as credible, versus a control group that was not exposed to AI-generated news site.

Further, trust in nonprofit organizations and in AI-generated news as well as social capital and civic engagement levels were all found to increase with increased exposure to the AI-generated news.

The site emerged as a part of the AIR group’s project on AI News, Civic Engagement, and Trust (ACT), an effort with a three-pronged goal of “promoting trust among individuals, reinforcing the credibility of journalism, and instilling confidence in nonprofit entities.”

The project prompts further exploration of how AI-generated news will continue to intersect with local journalism. Presentation of AI-generated news can be key to building trust not just in the articles themselves, but in the greater perception of a community and its institutions.

Under the study’s findings, perhaps AI news supplementing the human method of news production could pose a symbiotic relationship rather than a tug-of-war.

Authors: Seungahn Nah, Ernest Makata, Xinlei Wu (University of Florida College of Journalism and Communications), Jun Luo (University of California, Los Angeles), Seungbae Kim and Ian Koratsky (University of South Florida).

This summary was written by Caleb Wiegandt.

Posted: October 24, 2024
Insights Categories: AI, Communication and Technology
Tagged as: AIatUF, Artificial Intelligence, Consortium on Trust in Media and Technology, Ernest Makata, Seungahn Nah, Trust, Xinlei Wu