Ticker

6/recent/ticker-posts

Ad Code

Responsive Advertisement

AI-Generated “Impossible Statue” Is a Dispassionate Steel Mess

The “Impossible Statue” on display in Stockholm (photo by Anna Gerdén, courtesy National Museum of Science and Technology Stockholm)

As we collectively accelerate into the era of artificial intelligence, the powerful learning mechanism has yielded a number of novelty-factor takes on art, photography, screenwriting, and even movie trailers. The latest such “development” is something called the “Impossible Statue” — an allegedly first-of-its-kind 3D sculpture conceived by a generative AI that is based on the synthesis of styles from five legendary human sculptors from art history. The five-foot, half-ton stainless steel artwork is now on display at the Tekniska Museet, the Swedish National Museum of Science and Technology in Stockholm.

The sculpture design began with 2D generative AI tools including Stable Diffusion, DALL-E, and Midjourney, which were trained to incorporate the five artists’ areas of expertise: the dynamic off-balance poses of Michelangelo; the musculature and reflectiveness of Auguste Rodin; the expressionist feeling of Käthe Kollwitz; the focus on momentum and mass embodied in the work of Takamura Kotaro; and the defiance in the figures of Augusta Savage. The use of existing 3D AI interpreters, as well as new proprietary tools developed by Sandvik Machining Solutions in collaboration with a consulting firm known as the AI Framework, helped take all this learning from the mind of AI into our dimensional reality.

The result is … you know, fine. It’s a muscular and dispassionate stainless-steel figure draped in a flowing, pleated garment, and palming a golden globe like a basketball. Would anyone be impressed with this sculpture if it weren’t made by AI? No. Is it a better stainless-steel statue than I could make? Pretty much. Would it make a great, very high-end hood ornament for Elon Musk’s next venture into space? It probably will. Is it good art? No, but so aren’t lots of sculptures made by actual humans. And like all those human artworks, the “Impossible Statue” has attracted a flotilla of haters.

Good news for “Impossible Statue” haters: Like many high-profile AI art projects, it is less about creating tasteful or thought-provoking art and more a press-friendly way to broadcast proof of concept for a company’s prowess as an AI-enabled global manufacturer. According to Sandvik, the entire process also reduced the amount of steel needed for the artwork by half, which is nice to hear in a world grappling with the consequences of manufacturing excess, though obviously, massive corporations and not artists are responsible for such a crisis.

What does it mean for art? Maybe we should ask ChatGPT — after all, it might take an AI critic to fairly assess an AI sculpture in context. From this human perspective, it’s a little sinister that the figure holding our Earth seems less reverent and more about to mic-drop it. We’ve entered our post-human era, Earth artists. Go get a real job. You’ve been warned.

Sandvik cobbled together several 2D and 3D generative AI products, as well as some proprietary tools, to create AI that could interpret and translate the concepts into physical creations. (image courtesy Sandvik)

Enregistrer un commentaire

0 Commentaires