Midjourney, the popular artificial intelligence image-generator, has initiated a temporary measure to block its users from creating fake images of President Joe Biden and former President Donald Trump leading up to the U.S. presidential election. The decision came after tests conducted by The Associated Press revealed the potential for misuse of the tool. Midjourney CEO David Holz explained that the company wants to take a stand against election-related misinformation, but he did not provide specific details on the policy changes being implemented.
The move by Midjourney, which only employs 11 individuals according to its website, differs from the actions of other leading image-generating tools. The company did not participate in a voluntary industry pact that aimed to combat AI-generated deepfakes intended to deceive voters. Holz clarified that Midjourney’s primary purpose is not related to political speech, but the company also wants to avoid spending excessive time on policing such content.
Critics have raised concerns about the potential for Midjourney’s technology to fuel disinformation campaigns and false claims of election fraud. The Center for Countering Digital Hate recently published a report indicating that Midjourney has already been utilized to create images that could support such disinformation. Callum Hood, the head of research at the organization, highlighted that Midjourney stood out as the AI image-generator with the least controls when it came to producing images of well-known political figures like Biden and Trump. The watchdog group also tested other tools, including OpenAI’s ChatGPT Plus, Stability AI’s DreamStudio, and Microsoft’s Image Creator, all of which demonstrated some level of problems in generating election-related disinformation.
Despite Midjourney’s decision to block the creation of Trump and Biden images, there have been instances where users successfully produced such content in the past. However, the company’s public database indicates that this has changed recently, with users reporting warnings when attempting to generate images of the two political figures. Some users have expressed frustration over a lack of clear communication from Midjourney regarding its revised policy.
One aspect that sets Midjourney apart from other image-generators is its public nature. Users can view each other’s prompts and images on the social media platform Discord, where Midjourney’s largest group, comprising more than 19 million members, is located. CEO David Holz regularly interacts with users in this group, delving into discussions on a wide range of topics related to AI and the future of humanity.
Although Midjourney’s temporary blocking may address concerns about AI-generated images for the upcoming election, Holz predicts that the landscape will change significantly in the future. He believes that by 2028, all candidates will have realistic “deepfake chatbots” that simulate talking points. Holz warns that individuals who are determined to create deepfakes will find alternative methods that can be customized for specific targets and prove more effective than existing systems.
Established in 2020 and headquartered near San Francisco, Midjourney positions itself as an independent research lab focused on exploring new forms of thought and expanding the imaginative capabilities of humans. In July 2022, the company introduced the public version of its image-generator, generating significant interest in the field of generative AI technology. Despite striving to allow users the freedom to create a wide range of images, Midjourney actively blocks offensive and harmful content. The company employs keyword filters and maintains a team of moderators who address complaints from users.
FAQ
Q: Why has Midjourney decided to block the creation of AI-generated images of Biden and Trump?
A: Midjourney has implemented a temporary measure to prevent the creation of fake images of the two political figures in light of concerns surrounding election-related misinformation.
Q: How does Midjourney’s response compare to other image-generating tools?
A: While Midjourney has taken steps to address potential misuse of its technology, it did not join a voluntary tech industry pact aimed at combatting AI-generated deepfakes. Other leading image-generating tools also faced challenges in preventing the generation of election disinformation.
Q: What criticisms have been made against Midjourney?
A: Critics argue that Midjourney’s image-generator can contribute to disinformation campaigns and false claims of election fraud.
Q: What is unique about Midjourney’s approach compared to other image-generators?
A: One distinctive feature of Midjourney is that its platform operates in public forums on the social media platform Discord, allowing users to view each other’s prompts and images.
Q: What does Midjourney’s CEO predict for the future of fake images?
A: Midjourney’s CEO foresees a future where all candidates will have realistic “deepfake chatbots” that simulate talking points, making the creation of fake images even more sophisticated.
Definitions:
– AI image-generator: A tool or software that uses artificial intelligence to generate images.
– Deepfakes: Artificially manipulated or synthesized media, typically images or videos, that depict events or situations that never actually occurred.
FAQ:
Q: Why has Midjourney decided to block the creation of AI-generated images of Biden and Trump?
A: Midjourney has implemented a temporary measure to prevent the creation of fake images of the two political figures in light of concerns surrounding election-related misinformation.
Q: How does Midjourney’s response compare to other image-generating tools?
A: While Midjourney has taken steps to address potential misuse of its technology, it did not join a voluntary tech industry pact aimed at combatting AI-generated deepfakes. Other leading image-generating tools also faced challenges in preventing the generation of election disinformation.
Q: What criticisms have been made against Midjourney?
A: Critics argue that Midjourney’s image-generator can contribute to disinformation campaigns and false claims of election fraud.
Q: What is unique about Midjourney’s approach compared to other image-generators?
A: One distinctive feature of Midjourney is that its platform operates in public forums on the social media platform Discord, allowing users to view each other’s prompts and images.
Q: What does Midjourney’s CEO predict for the future of fake images?
A: Midjourney’s CEO foresees a future where all candidates will have realistic “deepfake chatbots” that simulate talking points, making the creation of fake images even more sophisticated.
For more information on Midjourney, you can visit their website: Midjourney
The source of the article is from the blog lanoticiadigital.com.ar