The capabilities of generative artificial intelligence are growing rapidly, and so is the frequency of AI-generated content. The increase in realistic yet fabricated content creates serious concerns regarding the information ecosystem and spread of disinformation.
Roger Stahl, a University of Georgia professor who studies propaganda and media systems, said the current disinformation we see is a preview of something much larger.
“We’re kind of walking into this house of mirrors,” Stahl said, “and we’re just seeing the very, very edge of it.”
Artificial intelligence has made producing realistic fabricated images, videos and text faster and more accessible than ever before.
Stahl traces the roots of today’s problem to social media’s rise around 2010, when information shifted from centralized outlets like television and newspapers to a fragmented landscape of platforms and influencers. That shift, he said, didn’t just change how news travels — it eroded trust in the institutions that once controlled the flow of information.
There’s a kind of limbic, instinctual trust in whatever personalities people happen to be following at the time,” Stahl said.
Audiences no longer turn to news organizations with editorial and ethical standards. Instead, credibility has become tied to identity, social circles and whoever someone follows online.
AI accelerates that dynamic by flooding the Internet with synthetic content that appears credible enough to spread.
“Authorship becomes a lot less clear, transparent and maybe less important,” Stahl said.
Kyle Johnsen, a professor in UGA’s College of Engineering and associate director of UGA’s Institute for Artificial Intelligence, sees the technical threat as both real and underappreciated. The same capabilities that power chatbots and image generators have made creating disinformation nearly effortless.
“Your knowledge about AI, if it is more than three months out of date, you might think AI is not capable of something, but it actually is,” Johnsen said.
Detection tools do exist and are improving, Johnsen said, but they only work if someone is actively looking.
“Most people can still tell the difference between an AI-generated photo and a real one, but it’s not something you’re just going to accept,” Johnsen said. “You’d have to be looking for it.”
One proposed solution — mandatory labeling of AI-generated content — has gained traction in Europe, where the EU AI Act includes transparency requirements for synthetic media. Stahl called it a meaningful starting point, while Johnsen said he didn’t believe this to be a realistic solution.
Johnsen said while policy frameworks matter, enforcement is the key, especially for bad actors who have every incentive to ignore labeling requirements.
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,” Johnsen said.
Stahl’s longer-term answer circles back to journalism itself. Established news organizations, he said, still have something bad actors don’t: accountability.
“There’s a profoundly fractured audience that are kind of insulated in their own little enclaves. And the move from a centralized media legacy system to social media system necessarily entails the delegitimization of the old system,” Stahl said. “The answer has always been established media organizations — organizations that have credibility and clout and have something to lose.”
Annabelle Foley is a journalism major and in the Certificate in News Literacy in the Grady College of Journalism and Mass Communication.
