Shulou(Shulou.com)11/24 Report--

Thanks to CTOnews.com netizen for the clue delivery of rain and snow! CTOnews.com July 9 news, in recent years, artificial intelligence (AI) in the creation of art has shown amazing ability. Just type a sentence into a text box and the AI can generate a beautiful image according to the instructions. The beauty of this technology lies in its ability to use human language to control the process of artistic creation, but do these systems accurately reflect the artist's imagination? Can integrating language into artistic creation really lead to artistic breakthroughs?

Playform AI, an AI art creation platform, conducted a survey that collected responses from more than 500 digital artists, traditional painters, photographers, illustrators, and graphic designers using platforms such as DALL-E, Stable Diffusion, and Midjourney. The results showed that only 46% of respondents found such tools "very useful," while 32% found them useful but unable to integrate them into their workflow, and 22% did not find them useful.

CTOnews.com notes that artists and designers point out that the biggest limitation of generative AI is lack of control, on a scale of 0 to 10, with 10 indicating maximum controllability, and respondents indicating their ability to control outcomes between 4 and 5. Half of respondents found the results interesting, but not of high enough quality to be used in practice.

When it comes to whether generative AI will affect their creative practices, 90% of artists surveyed believe it will, with 46% believing it will be positive, 7% believing it will be negative, and 37% believing it will be affected, but not sure how.

Some artists believe that the biggest limitation of generative AI in terms of art is that language is the main driving force for generating images. Visual artists are by definition visual thinkers, and when they imagine their work, they usually base it on visual references rather than words. When language becomes the focus of image creation, it blocks the link between the artist and the digital canvas, and the artist loses the freedom to manipulate pixels outside the semantic realm.

Another fundamental limitation is the uncertainty of the results, since if two artists enter exactly the same cue word, the system will most likely not produce the same image. In other words: the artist's work is limited to chance.

Text-to-image technology is an interesting new tool that allows users to generate images quickly. But for artists, this tool is not a creative solution, it limits the artist's control, freedom and identity expression. The best visual art transcends language and is not bound by language.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.