AI generated images
AI-generated images of Joe Biden and Ron DeSantis in compromising situations. The person who created these fake images made them to show how the technology can be used as a weapon.

An artist who created AI images of politicians cheating on their spouses did so to warn of the dangers of artificial intelligence (AI) — but he was banned from Midjourney for it.

Video editor and political satirist Justin T. Brown used Midjourney to generate a series of fake images of political figures such as Donald Trump, Barack Obama, and Joe Biden cheating on their partners.

Brown, who is based in Los Angeles, says the hyperrealistic images took him a little over an hour to make on Midjourney.

The image series — which is entitled “AI will revolutionize the blackmail industry” — was intended to show how the technology could be used as a weapon in the wrong hands and create scenarios that could hurt people.

“I created the series with the goal of dragging some of the most influential people in America into the conversation about the dangers of AI, and reach the biggest possible audience,” Brown tells PetaPixel.

“The point was to create buzz and spark conversation in the most salacious, shocking, and funny way possible.

“In order to show the real potential threat of AI, the project itself needed to be a threat, and to my mind there’s no safer target than powerful men in Washington. It was intended to be a shot across the bow.”

AI generated images
Fake AI-generated images of Donald Trump and Barack Obama in compromising situations.

Brown — who never presented the images as being real — shared the AI photo series on the Midjourney subreddit.

Brown had created the images on Midjourney months before. However, within two days of the photos being posted on Reddit, Brown promptly received a ban from Midjourney.

“After gaining some traction on Reddit, the series was removed by moderators and the Midjourney ban followed almost immediately,” he explains.

“I’ve come up against blocked prompts in the past — for naughty words or controversial figures — but never received a ban.

“I wasn’t given a direct reason for the ban by Midjourney, only a message saying I was blocked, but the timing of the Reddit release and the ban correlate directly. That’s a space where Midjourney folks hang out.”

‘I Feel Every Possible Way About It’

In a viral Twitter thread, which received over one million views, Brown discussed his conflicting feelings about getting banned from Midjourney and how easily fake yet believable images could be generated with AI.

“I feel every possible way about it. I’d rather not be banned, but at least maybe that means there’s some accountability for when users generate harmful content with bad intent,” Brown explains.

“But I was banned months after I first made the images, so by the time harmful content is generated, isn’t it already too late?

“Some regulation of content is necessary I think, but why would I ever trust a company to regulate its own bottom line? And who gets to decide what kind of content is to be regulated?”

‘Devastating Potential’

Brown says that one of his major concerns with AI is its ability to spread disinformation and that there needs to be an “international conversation” about the technology.

“I never presented the images [in the photo series] as real. But I do know that some of those images have been taken and shared out of context by others, so I wouldn’t be surprised if there are some people who believed they were real,” Brown says.

“But to some extent, that’s part of the project. This kind of disinformation will continue to happen, next time by bad guys with bad intent, so we better start getting a handle on it now.”

He adds: “In a better world, I’d welcome AI. But in the world we live in, it has the potential to be devastating. I think our world has to change.”

A report published by The New York Times this weekend revealed that the gaps in campaign rules have enabled politicians to disseminate images generated by AI in the lead up to the upcoming 2024 presidential election,

And existing defenses, such as social media rules and services that claim to identify AI content, have failed to do much to slow the tide.

More of Brown’s work can be seen on Twitter, YouTube, and Instagram.

Image credits: All photos by Justin T.Brown.

Read More