People were ecstatic about these programs since the outcome was stunning and they could be used by everyone. I’ve been toying with these tools and became interested in constructing visualization models with AI that unravel how the mind processes information, and I can vouch to their democratizing impact. I’ve created hundreds of photographs almost completely for the joy of gazing at them. My friend once completed a Pac-Man-themed painting series that I was rather pleased with.
The art world has been turned upside down by the ease of use of picture programs and the appealing outcomes. Professional artists are concerned about losing work to anybody with a computer and the ability to follow a few instructions. Artists working for video game firms have already reported that smaller shops have ceased recruiting as many employees for their creative requirements. Illustrators are fuming as publishers use AI-generated graphics instead of their work. Artists have my sympathies. After spending decades honing your abilities, it must be demoralizing and terrifying to confront a computer that can handle elements of your work at a fraction of the cost. At the same time, the democratizing powers of modern technologies have the potential to create a more diverse creative environment for all of us.
True, AI builds an image in a very different manner than humans do. Diffusion models, such as Disco Diffusion, Stable Diffusion, and Midjourney, begin with random noise and gradually transform the noise to seem more and more like what it “thinks” a picture should look like in response to a suggestion. Humans, on the other hand, often begin with a high-level notion of what the picture’s features will be, what viewpoint it will take, and so on, and then execute the specifics. But does the audience care about this distinction?