ALBAWABA – With the rise of Generative Artificial Intelligence, allowing anyone to be able to create images of celebrities that are fake or outright embarrassing, like the wave of inappropriate Taylor Swift images that circulated last month, concerns about AI use to discredit politicians in the upcoming US election are causing many to worry, including Midjourney, an advanced AI software that allows creation of hyper-realistic images, whose CEO, David Holz, mentioned the firm is considering a ban on creating images of Trump and Biden ahead of the elections.
In a Discord session with Midjourney users, Holz said “I know it's fun to make Trump pictures — I make Trump pictures,” adding that while Trump has an intriguing look that motivates creation of fake AI images, he advised to abstain from that at least for the election period according to Bloomberg, noting that a ban on such figures for the next 12 months is being thought of.
Midjourney had earlier discontinued their free membership tier after fake images of the pope dressed in a luxurious jacket became trending, as well as fake images of Trump’s arrest and mugshots created with the company’s software circulated around the internet.
AI fake of The Pope that went viral and was assumed real by many... pic.twitter.com/LeHn3SdAAD
— David Baszucki (@DavidBaszucki) February 8, 2024
Other platforms are also jumping on board in fighting political misinformation caused by generative AI, with Meta announcing earlier that it will tag AI content with a software that works on identifying fake posts and labelling them so no one is fooled, and Open AI, the company behind popular ChatGPT, announcing their plans to watermark images created with their Dall-E generative AI software.
