On the heels of Facebook’s announcement it would ban AI-driven deepfakes, but not traditionally created shallow fakes, Reddit has announced a broader and more nuanced policy that demonstrates far more common sense.
Facebook’s policy has already been criticized for not doing enough to combat deepfakes, where images or videos of a person can be manipulated or even superimposed in an effort to show them saying or doing something they are not. While AI-enabled deepfakes are the most convincing, and almost impossible to detect, traditionally generated shallow fakes can be almost as convincing. Facebook’s failure to address the latter leaves the social media platform open to further manipulation.
Reddit, on the other hand, is focusing more on intent than a legalistic “do this, don’t do that” approach. A post on the site laid out their approach:
“ Do not impersonate an individual or entity in a misleading or deceptive manner.
“Reddit does not allow content that impersonates individuals or entities in a misleading or deceptive manner. This not only includes using a Reddit account to impersonate someone, but also encompasses things such as domains that mimic others, as well as deepfakes or other manipulated content presented to mislead, or falsely attributed to an individual or entity. While we permit satire and parody, we will always take into account the context of any particular content.”
As the above statement indicates, satire and parody are safe. Another part of the post makes that even clearer: “This doesn’t apply to all deepfake or manipulated content– just that which is actually misleading in a malicious way. Because believe you me, we like seeing Nic Cage in unexpected places just as much as you do.”
Reddit’s approach, especially compared to Facebook, is a refreshing, common sense approach the other social media platforms would do well to follow.