“Given the creativity humans have shown throughout history to invent (false) stories and the freedom they already have to create and spread disinformation across the world, it is unlikely that a large portion of the population is looking for misinformation that they can’t find online. or offline,” the newspaper concludes. Additionally, misinformation only gains momentum when people see it, and given that the time people have to view viral content is limited, the impact is negligible.
When it comes to images that might end up in mainstream feeds, the authors note that while generative AI can theoretically render highly personalized and very realistic content, so can Photoshop or video editing software. Changing the date on a grainy cell phone video could be just as effective. Journalists and fact-checkers have less difficulty with deepfakes than with out-of-context images or those crudely manipulated to make them into something they are not, such as video game sequences presented as a Hamas attack.
In this sense, Excessive focus on flashy new technology is often a red herring. “Being realistic is not always what people are looking for or what it takes to go viral on the Internet,” adds Sacha Altay, co-author of the paper and a postdoctoral researcher whose current field concerns misinformation, trust and social media at the University. from the Digital Democracy Lab in Zurich.
This is also true on the supply side, says Mashkoor; the invention is not the implementation. “There are many ways to manipulate the conversation or manipulate the information space online,” she says. “And there are things that are sometimes less difficult or easier to do and may not require access to specific technology. Even though AI generation software is easy to access at the moment, it There are definitely easier ways to manipulate something if you’re looking for it.
Felix Simon, another author of the Kennedy School paper and a doctoral student at the Oxford Internet Institute, cautions that his team’s comment does not seek to end the debate over possible harm, but rather is an attempt to push back people’s statements. AI will unleash “an armageddon of truth.” This kind of panic often accompanies new technologies.
Putting aside the apocalyptic vision, it is easier to study how generative AI has In fact inserted into the existing disinformation ecosystem. It is, for example, much more widespread than it was at the start of the Russian invasion of Ukraine, says Hany Farid, a professor at the School of Information at UC Berkeley.
