Menu Menu

AI-generated piece controversially wins art competition

This year at the Colorado State Fair, a piece of AI-generated art won an award for Emerging Digital Artists. In the days since, online debate about what constitutes ‘real art’ has ignited.

If you keep up with Thred, you’ll already be completely in the loop regarding AI-generated ‘text-to-image’ programs and their recent emergence.

Canny visual software like DALL-E, NightCafe, and – in this instance – Midjourney, are able to take our words as prompts and transform them into detailed (and often quite original) compositions.

Only recently, we covered a story where amateur city planners created digital mockups of famous city hotspots on DALL-E, and reimagined what they would look like if fully pedestrianised.

Despite being previously locked down by giant corporations like OpenAI and Google, these programs have since become easily accessible and are constantly used to create meme material for social media.

While, for the most part, text-to-art generators have been used as tools of recreation, recent events show the potential for AI works to leave a lasting impact on the professional scene too. Though, not everyone is championing this notion.

At the Colorado State Fair last week, its annual art competition awarded prizes in all the typical categories: painting, sculpture, quilting, etc. One winning entry, however, contentiously took first prize in the digital art rankings for an AI piece created using Midjourney.

The artist, Jason Allen, a studio head at a tabletop games company called Incarnate, used the Discord based text-to-image generator to create the work titled ‘Theatre d’Opera Spatial.’

The piece itself is certainly very striking. Women adorned in renaissance style robes marvel at a futuristic landscape through a giant circular visor. Meanwhile, natural light shines through revealing intricate details on the oily textured surfaces and obscure displays.

This digital composition was one of many Allen had created over a matter of months, and he decided to blow it up on canvas to be sent to the fair. ‘I thought: How wonderful would it be to demonstrate to people how great this art is?’ he said to the New York Times.

Upon winning, he was obviously jubilant, but there has since been significant backlash from peers and enthusiasts – much of it, predictably from the Twittersphere – who assert that AI work lacks integrity and artistic merit.

‘We’re watching the death of artistry unfold right before our eyes,’ one Twitter user proclaimed.

‘This is so gross,’ said another. ‘I can see how AI art can be beneficial, but claiming you’re an artist by generating one? Absolutely not.’

For many who earn a crust in the industry, there is understandable trepidation about the very nature of text-to-image applications. Any time an artist uploads their work online, they are inadvertently improving the quality of the technology.

As the AI becomes more sophisticated, it will only become harder to distinguish between purely original pieces and digitally spliced ones too. This grey area is highlighted by the fact that Allen’s judges reportedly awarded him the winning spot without knowing an AI had been involved at all.

Then again, the category’s rules allowed for any ‘artistic practice that uses digital technology as part of the creative or presentation process.’ Technically, there’s no foul play then.

To those who point to a ‘lack of work’ in attaining the final product, Allen snaps back stating that prompts were fine tuned for weeks, images painstakingly curated, and general composition work completed on Photoshop.

Whatever people’s personal preference on this technology entering the industry, it appears that text-to-image software isn’t going anywhere just yet.

Allen ominously declares: ‘This isn’t going to stop. Art is dead, dude. Its over… AI won; humans lost.’ Spoken like a true artist.