Generative AI technology enables users to create a variety of synthetic images, including deep fakes, which have recently featured former President Donald Trump and Russian President Putin in various humorous and absurd contexts.
Making pictures of Trump getting arrested while waiting for Trump's arrest. pic.twitter.com/4D2QQfUpLZ
— Eliot Higgins (@EliotHiggins) March 20, 2023
However, one notable exception is the Chinese President Xi Jinping, as the generative AI platform Midjourney has banned the creation of his image or even the mention of his name as a prompt.
This move has drawn criticism from some, who see it as a form of censorship that goes against the principles of free speech and expression.
Sarah McLaughlin, a Senior Scholar at the Foundation for Individual Rights and Expression (FIRE), has argued that such censorship may have wider implications beyond the realm of generative AI technology. The actions of companies like Midjourney may be influenced by the speech laws of other countries, such as China, which could impact how American companies operate. McLaughlin cited the example of the 2015 Sony hack, which revealed that a scene in the movie “Pixels” depicting the Great Wall of China being destroyed was removed to avoid offending the Chinese government and jeopardizing the film’s box office success in China.
McLaughlin also emphasized the importance of not dismissing international censorship as irrelevant, as it may have unforeseen consequences in the future. While Midjourney CEO David Holz has defended the ban on images of President Xi as a way to minimize drama and avoid potential problems in China, McLaughlin sees it as a complex issue that could ultimately lead to the restriction of free speech and expression in other countries.
However, supporters of such restrictions argue that they are necessary to prevent the spread of disinformation and the potential misuse of AI technology.
Despite the controversy, McLaughlin does not see Midjourney’s time in China lasting very long, as it may ultimately be blocked by the Chinese government, as is common with Western companies. She also highlighted the differences between Midjourney’s approach and that of Twitter, which censored content in India only in response to government demands, rather than imposing restrictions on all users.
McLaughlin’s comments underscore the complex and nuanced nature of censorship and free speech in the digital age, and the potential implications for individuals, companies, and governments around the world.