Instagram “photographer” known as Joe Avery, who has almost 37,000 followers on his Instagram photography account, admitted his “masterpieces” are AI-generated.
His account @averyseasonart contains black and white portraits that garnered praise in the comment sections. However, none of the people in the pictures are real.
The artist has posted 185 “portraits” alongside the subjects' names and an anecdote or story about their lives. However, the portraits were all generated using artificial intelligence software and basic Photoshop skills. Avery admitted he has been using Midjourney, an AI image generator, and then fine-tuning the images using Photoshop.
Joe Avery considered his portraits “AI-generated human-finished portraits.” However, none of the photographs were acknowledged as such on his Instagram page. He previously said his pictures were taken using a Nikon D810 camera.
After many followers noticed the portraits were AI work, fans asked him to mention it on his Instagram bio.
“I created this image using an AI-generated image, digital drawing, and editing in Photoshop,” he answered one of his fans' questions about the lighting. According to Euronews, the artist admitted the popularity of the images, which he began posting on Instagram in October 2022, had wildly exceeded his expectations and caused him to feel uneasy about the truth behind them.
The Instagram account has blown up to nearly 12K followers since October, more than Avery expected. “Because it is where I post AI-generated, human-finished portraits. Probably 95%+ of the followers do not realize. I’d like to come clean,” he told Ars Technica.
Although many followers expressed their admiration towards the pieces, others criticized the artist’s work. “skimming through your posts, though you do seem to be disingenuously misleading people from the start…This seems very intentionally misleading,” an Instagram user commented. It remains to be seen whether he will face backlash from his followers, websites suggested.
Joe Avery’s initial aim was “to fool people by showing how impressive AI could be, then write an article revealing the truth,” Euronews Culture reported. However, the artist decided to show that his photographs required effort. “Now, it has become an artistic outlet. My views have changed,” he told Ars Technica.
For the artist, these images are still art, produced in a different way. The process also takes time generating thousands of pictures in Midjourney, and selecting the best to put through editing in both Lightroom and Photoshop.
With the rise of AI software use, many artists face similar situations, especially those using AI within their work. Those artists receive backlash from their fans and pushback from more “traditional” artists. However, there are many competitions and awards for AI-generated visuals.
Last month, an award was given to an AI-generated picture in a photography competition in Australia after the judges of the competition were completely fooled, believing the image to be a real photograph taken by a drone of surfers on a beach at sunset.
Artificial Intelligence Can Spread Misinformation
As AI mimics the process of creating content and sharing it on social media, media’s reliance on this new technology can have unexpected consequences. Fake photos of people who never existed can be produced, and videos simulating real-life movements and speech can be generated.
For Professor Afafe Annich of the Higher Institute of Journalism and Communication ISIC in Morocco, artificial intelligence human behavior simulation on multiple levels. According to Annich, artificial intelligence is based on learning and acquiring knowledge that will later be applied in various situations.
"Artificial Intelligence allows replicating humans’ capabilities and performing tasks requiring human intelligence," she explained.
Professor Annich previously explained to Misbar that while the AI process may appear simple to the general public, it may require several complex stages and software to perform a simple task using an AI tool. She also mentioned that large datasets could help in the machine-learning process.
However, Artificial Intelligence can also defeat misinformation. Since artificial intelligence produces misleading content, debunking the false content requires a reverse process. The process may also be based on identifying the altered or fabricated photos or identifying any information that will help debunk the content.
Misbar has recently reported on Venzuela’s new weapon: AI-generated propaganda. The process was based on using computer-generated avatars to spread positive news about the country through a YouTube channel called “House of News.” According to our blog, these avatars are entirely fake and created by London-based AI company “Synthesia.”
Similarly, Artificial Intelligence can allow people to invent a social life by creating images that suggest they attended a social gathering that never happened and with friends that do not exist. Photos of women smiling at the camera and men raising toasts during a party made round on social media. Many may say the portraits look realistic. However, “people” in the pictures were created with more than five fingers on their hands and mouths full of teeth. These portraits were also generated by the AI system Midjourney.