UGA Experts: Generative AI is Accelerating the Erosion of Public Trust

Experts from the University of Georgia are warning that the rapid advancement of generative AI is flooding the information ecosystem with realistic but fabricated content. This surge in synthetic media makes the creation of disinformation nearly effortless, further destabilizing the public's ability to identify truth.

Professors Roger Stahl and Kyle Johnsen, representing an interdisciplinary effort between the University of Georgia's Institute for Artificial Intelligence and the Grady College of Journalism and Mass Communication, highlight that generative AI capabilities and the frequency of AI-generated content are growing at a rapid pace. They argue that while detection tools and labeling policies exist, these measures are often insufficient when facing determined bad actors. "We can’t just put policies out there and hope for the best, especially when we’re dealing with something that is really, really difficult," said Professor Johnsen.

The Path to Fragmentation

This crisis is the culmination of a decade-long shift in how information is consumed. According to the experts, the transition from centralized media to fragmented social media around 2010 began the erosion of trust in traditional institutions. Generative AI now acts as an accelerant to this existing trend, removing the barriers to entry for producing high-quality, deceptive media at scale.

Professor Stahl described the current information landscape as a "house of mirrors," noting that the public is currently seeing only the "very, very edge" of the potential impact. This environment makes it increasingly difficult for users to rely on visual or auditory evidence as a proxy for truth.

Implications for the Information Ecosystem

The ability to produce high-quality fabricated media at scale threatens the fundamental stability of the information ecosystem. As credibility shifts away from established editorial standards and toward social identity and influencers, the lack of transparent authorship becomes a critical vulnerability. When the origin of a piece of content is obscured, the public is left without a reliable mechanism to distinguish verified reporting from AI-generated disinformation.

This shift places a heavier burden on the individual consumer to verify sources in an environment where the tools for deception are evolving faster than the tools for detection.

The Regulatory Landscape

Efforts to combat this trend are emerging through legislation, though their effectiveness remains to be seen. The EU AI Act, for example, now includes transparency requirements that mandate the labeling of synthetic media to ensure users know when they are interacting with AI-generated content.

However, the UGA experts suggest that regulatory frameworks alone may not be enough to counter the speed of AI development. The primary challenge remains the gap between the ease of creating synthetic content and the difficulty of verifying it in real-time across fragmented social platforms.