In just one Over the past few years, the number of artworks created by self-described AI artists has increased dramatically. Some of these works have been sold by great auction houses with dizzying prices and found the way in curated prestigious collection. Originally led by a number of tech-savvy artists who used computer programming as part of their creative process, AI art has recently been embraced by the masses, as visualization technology becomes more efficient and easier to use without coding skills.
The AI art movement is based on a combination of technical advances in computer vision, a field of study dedicated to the design of algorithms that can process meaningful visual information. A subclass of computer vision algorithms, called generative models, takes center stage in this story. Generative models are artificial neural networks that can be “trained” on large data sets containing millions of images and learn to encode their statistically salient features. After training, they can generate entirely new images that were not present in the original dataset, often guided by text prompts that clearly describe the desired outcome. Until recently, the images created using this approach were somewhat incoherent or lacking in detail, although they possessed an undeniable surreal appeal that attracted the attention of many artists. serious. However, earlier this year, technology company Open AI unveiled a new model – nicknamed DALL · E 2—That can produce remarkably consistent and consistent visuals from virtually any text prompt. DALL · E 2 can even create images of specific styles and imitate famous artists quite convincingly, as long as the desired effect is fully specified in the prompt. A similar tool has been released to the public free of charge under the name Craiyon (formerly “DALL · E mini”).
The coming age of AI art raises some interesting questions, some of which – such as whether AI art is really artisticAnd if so, to what extent? made by AI—Not particularly original. These questions echo similar anxieties that were raised when photography was invented. With the push of a button on the camera, a person with no art skills can unexpectedly capture a realistic depiction. Today, a person can press a virtual button to run a generalized model and create an image of virtually any scene in any style. But cameras and algorithms don’t make art. Everyone does. AI art is art, created by human artists using algorithms as another tool in their creative arsenal. While both technologies have lowered the barrier to entry into artistic creation – which calls for celebration rather than concern – one should not underestimate the amount of skill, talent and intent involved in making art. create interesting works of art.
Like any novelty tool, synthetic models introduce significant changes in the creation of artwork. In particular, AI art expands the multifaceted concept of curation and further blurs the lines between management and creativity.
There are at least three ways to make art with AI that can involve curatorial activities. The first, and least original, concerns the management of outputs. Any compositing algorithm can produce an indeterminate number of images, but not all of these images will generally be recognized as state of the art. The process of managing outputs is very familiar to photographers, some of whom often take hundreds or thousands of images from which a few photos, if any, can be carefully selected for display. Unlike painters and sculptors, photographers and AI artists have to deal with a multitude of (digital) objects, whose management is part of the artistic process. In AI research in general, the act of “picking cherries” of exceptionally good output is considered bad science practice, a way to falsely inflate the cognitive performance of a model. However, when it comes to AI art, cherry picking might be the name of the game. The artist’s intentions and artistic sensibilities can be expressed in the very act of promoting specific outcomes for the state of the artwork.
Second, management can also happen before any images are created. In fact, while “curator” applied to art usually refers to the process of selecting existing work for display, curation in AI research colloquially refers to the work that goes into creating a dataset to train an artificial neural network. This work is important, because if a data set is poorly designed, the network will often fail to learn how to represent the desired features and perform adequately. Furthermore, if a data set is biased, the network will tend to reproduce, or even amplify, that bias — including, for example, harmful biases. As the saying goes, “garbage in, garbage out.” The adage holds true for AI art as well, except that “junk” has an aesthetic (and subjective) aspect.
