Firms' desire for ad revenue tied to inadvertently financing online misinformation outlets

Companies and digital platforms contribute to financially sustaining misinformation outlets via advertising. Despite attempts to reduce misinformation, ads from well-known firms and organizations continue to appear on misinformation websites, thereby financing such outlets. The supply of falsehoods is expected to rise with artificial intelligence making it easier to create large volumes of misinformation to earn ad revenue.

In a new study, researchers examined the reasons behind the spread of online falsehoods. Based on their findings, they suggest interventions to reduce the financing of misinformation. Conducted by researchers at Carnegie Mellon University and Stanford University, the study is published in Nature.

“Online misinformation can have significant consequences, including sowing political discord and exacerbating the climate crisis,” notes Ananya Sen, assistant professor of information systems and economics  at Carnegie Mellon’s Heinz College, who coauthored the study. “Our work is a first step toward understanding how to limit the financing of online misinformation via advertising.”

In the study, researchers addressed three issues. First, to evaluate the roles of advertising companies and digital ad platforms in monetizing misinformation, they constructed large-scale data sets, combining data on websites that publish misinformation with ad activity per website from 2019 to 2021. Their data set included nearly 5,000 websites (approximately 1,250 of which were misinformation websites) and more than 42,000 unique advertisers, with more than nine million instances of advertising companies appearing on news websites in the three-year period.

The study found that advertising on misinformation websites is pervasive for companies across several industries and amplified by digital ad platforms that use algorithms to distribute ads across the web. Misinformation websites are primarily monetized vis advertising revenue, with a substantial proportion of companies across several industries appearing on such websites, and the use of digital ad platforms amplifies the financing of misinformation.

Second, to measure consumers’ preferences, researchers conducted an experiment with a sample of the U.S. population by randomly varying the pieces of factual information provided to participants, then measured their reactions.

The study found that companies advertising on misinformation websites can face substantial backlash from consumers. Consumers switched away from companies whose ads appeared on misinformation outlets, reducing the demand for those firms’ products. The switching effect persisted even when consumers were informed about the role played by digital ad platforms in placing companies’ ads on misinformation websites and the role played by other advertising companies in financing misinformation. Consumers also voiced concerns about the practice, signing petitions advocating for firms to stop placing ads on misinformation websites.

Finally, to examine why misinformation continues to be monetized despite the potential for consumer backlash, researchers surveyed corporate decision makers to gauge what they know about misinformation online. While firm leaders said they believed most companies advertised on misinformation websites, they significantly underestimated their own company’s likelihood of doing so, the study found.

This suggests that many corporate leaders are ill informed about this possibility, and that firms may therefore be financing misinformation inadvertently, the authors say. Upon learning that their ads appeared on misinformation outlets, corporate leaders wanted to learn more and expressed interest in identifying platform-based solutions to reduce monetizing information.

Based on these findings, the authors propose two low-cost, scalable interventions to decrease the financing of misinformation:

• Improving transparency for advertisers about where their ads appear could reduce advertising on misinformation websites, especially among companies that were unaware of their ads appearing on such outlets. Information-based interventions could also be incorporated into existing legislation to improve transparency.
   
• While consumers can currently find out about ad companies financing misinformation through news and social media, platforms could make it easier for consumers to identify which companies advertise on misinformation outlets, for example, through simple information disclosures and comparative company rankings.

“Our findings have clear, practical implications,” suggests Wajeeha Ahmad, a Ph.D. student in management science and engineering at Stanford University, who led the study. “Given the potential for a substantial decline in consumer demand, ad companies may want to account for consumer preferences in placing their ads across various online outlets and exercise caution when incorporating automation in their business processes via digital ad platforms.”

Since consumer backlash was particularly strong for women and consumers who leaned left politically, companies targeting these audience may want to exercise greater caution.

The study was funded by the Stanford Digital Economy Lab, the Stanford McCoy Family Center for Ethics in Society, Stanford Impact Labs, Stanford Technology Ventures Program, Project Liberty, and the University of Pennsylvania’s Economics of Digital Services initiative.