Midjourney is set to launch an enhanced web tool that will enable users to edit any images uploaded from the internet using its generative AI technology. According to Midjourney CEO David Holtz, this upgraded tool is expected to be available “early next week. ” It will allow users to retexture elements within images, effectively "repainting" their colors and details based on text captions. The topic of AI-assisted image editing has ignited significant debate recently. Companies like Meta have faced challenges in determining how to label images altered using AI tools as opposed to those created entirely by AI. Meanwhile, Google has introduced powerful AI features that give no visual cues indicating that an image has been modified by AI technology. Last year, Midjourney pledged to implement the IPTC’s Digital Source Type property, a standard that embeds metadata in images to signify that they are AI-generated. However, it remains one of the few major AI platforms that has not adopted C2PA, which is a technology designed to trace the complete history of an image, including the tools and software used in its creation. Holz mentioned in a post on Midjourney’s official Discord server that the new image editing tool will initially be available only to a "subset of the current community, " with plans for greater human moderation alongside “new, more advanced AI moderators” to help mitigate potential misuse. “Honestly, we’re still figuring out how to accurately limit access to this feature, ” he noted. Midjourney is seeking input from the community through a poll to decide which users will gain priority access. There are significant risks associated with launching such editing tools without sufficient safeguards.

These tools could lead to widespread copyright violations or enable the dissemination of misleading deepfakes. Deepfakes have rapidly proliferated across social media, complicating the distinction between reality and misinformation. Recently, fake generative AI images depicting destruction and human suffering circulated online after Hurricane Helene. Data from Clarity, a deepfake detection company, indicates that there has been a 900% increase in the production and publication of deepfakes this year compared to the same period last year, causing understandable alarm. A recent YouGov poll revealed that 85% of Americans are worried about the online spread of misleading deepfakes. Without a federal law in the U. S. outlawing deepfakes, more than ten states have implemented laws targeting AI-assisted impersonation. California’s proposed legislation, currently stalled, would be the first to grant judges the authority to order the removal of deepfakes or potentially impose monetary penalties on their posters.