A sad threat to US politics – Bestgamingpro
Researchers at MIT’s Media Lab collaborated with Google’s Jigsaw group to investigate the impact deepfake AI political advertising could have on American voters.
In total, over 7,000 participants took part in the paired studies in the United States – this is, as far as we can tell, the most comprehensive study of its kind. Participants were divided into one of three groups: a deepfake video, a text transcription of the video, and none Media prompts.
In the third phase, participants were asked whether they agree with certain allegations or whether they believed the media they had seen, read or heard.
For those who want to lose weight, things don’t look any better, according to the study. Let’s start with the bad news. According to research:
Overall, we find that people are more likely to believe that an event has occurred when presented in video rather than text.
That might not be spectacular, but it turns out that what people see is far more trustworthy than what they read. Obviously, that’s a bad thing at a time when deepfakes are so easy to create. But it gets worse. According to a study:
Additionally, the difference between the video and text conditions in terms of attitudes and engagement is comparable, if not smaller, than the difference between the text and control conditions. Taken together, these results challenge popular beliefs about the unique persuasiveness of political video over text.
In other words, Americans are more likely to believe in deepfakes than text-based fake news, but they have little control over their political beliefs. Researchers hesitate to draw too many conclusions from these data.
They stress that the circumstances in which the study was conducted does not necessarily reflect those in which US voters are most vulnerable to deepfakes.
According to a study:
It should be noted, however, that while we observe little differences in persuasiveness between video and text in our two studies, the effects of these two modalities may diverge more widely outside of an experimental context.
In particular, it is possible for videos to be more attention grabbing than text so that people scrolling social media are more likely to pay attention and therefore be exposed to video versus text.
As a result, even if video has a limited persuasive advantage over text in a controlled setting with forced choice, it could still exert an overwhelming influence on attitudes and behavior in an environment in which it is disproportionately noticed.
Okay, deepfakes could be more successful in the wild when it comes to convincing people to change their political beliefs. But this particular study refutes that theory. From our point of view it seems logical.
In the last election, 50 million more Americans cast their votes than anyone else in US history. On the other hand, margins were so tight that one side is still (stupidly) claiming the election was stolen.
In fact, two of the four youngest US presidents lost the referendum. This suggests that American voters are not fickle.
Deepfakes are a relatively minor issue in US politics, as they suggest. It’s a little sad to see how easily our country’s partisanship has been defined by MIT research.
Subtle, charming pop culture freak. Amateur analyst. Freelance TV fan. Coffee lover