Some publications have begun testing the waters by publishing content generated by artificial intelligence, as public distrust of the news media has reached unprecedented levels, leading to disastrous outcomes in numerous cases.
This trend has led to a growing skepticism among users, further diminishing their trust in the content they consume.
A recent preliminary study by researchers from the Universities of Oxford and Minnesota revealed that readers are demanding transparency regarding the use of AI in content creation. Surprisingly, when news agencies disclose that articles are AI-generated and provide alternative options, readers are inclined to trust them less.
The implications of these findings are significant, as they shed light on how the disclosure of AI-generated content may impact market confidence in the news industry, especially at a time when its relationship with the public is fragile. The researchers emphasize that media organizations are increasingly turning to AI technologies in their editorial processes.
To conduct their research, the scholars surveyed 1,483 English speakers in the US, exposing them to a selection of AI-generated social media content, some labeled as such and others unlabeled. Participants were also presented with a set of news articles from various sources, some of which were AI-generated.
The participants were then asked to assess the credibility of the media outlets by reading the articles. The study revealed that articles labeled as “AI-generated” were rated lower on an 11-point trust scale compared to those without such labels.
Interestingly, respondents did not perceive the AI-generated reports as less accurate or more biased. Over 80% of the participants expressed the desire for news organizations to disclose whether the content was AI-generated.
The researchers acknowledged certain limitations of the study, including existing political divides and variations in media trust based on these divisions. The lack of real-world associations among the fictitious news entities mentioned in the study may have also influenced participants’ perceptions.
This complex issue underscores the necessity for further research, increased transparency, and meticulous scrutiny of AI-generated content by news outlets.
Coauthor and researcher Benjamin Toff from the University of Minnesota expressed optimism to Nieman Lab, suggesting that not all audiences may view the integration of AI technologies in newsrooms negatively. He highlighted the importance of framing these applications in a positive light to build trust rather than erode it.
