Are AI image-generation tools like Dall-E and Midjourney truly all-powerful and a threat to the livelihoods of trained artists?
In my experience, the tools themselves are somewhat repetitive and “dumb” with their generations unless you know how to work with them and create prompts that they can understand and emulate. This means that it isn’t as big of a threat to trained artists as I previously believed, but that doesn’t completely diminish artists’ fears. I will show examples of a bad prompt and its result vs. a good prompt. I’d also like to show an example of how AI can completely rip off a specific artist’s style to show how it can be threatening. Talking about how I had a hard time (and still do) speaking the AI tools’ languages will provide context to my experiences. It’s hard for me to make prompts with my existing, non-AI-related vocabulary. (The tools cannot always understand and produce work properly to the standards and quality of my vision.)
For the essay, I want to textually explain the creative question and my answer based on my experience with these tools and the class overall. For the final project, I’d like to heavily focus on how the quality of prompts is the difference between AI art generation tools being a threat to trained artists or not. I will give many examples, show how prompts evolve over time, and keep track of just how long (time-wise) it can take for an amateur like myself to generate the images that they envision. It might be cool to find an example online of an “expert” in these two tools and how their processes differ and how that affects the quality of their results as a comparison (if I’m allowed to do so, it will be properly cited and attributed, of course). I’d like to present the final project as an edited video with my creative process narrated with the help of ElevenLabs. It will include photo generations, text prompts, and anything else that I think would be helpful to showcase the answer to my creative question.