Is AI Art the Future of Web Images?¶
Source: https://www.wte.net/Is-AI-Art-the-Future-of-Web-Images
Date: October 2022
Author: Martin W Smith
Introduction¶
Code, images, and copy form the foundation of websites. Locating quality, affordable images remains challenging, making "text to art" AI tools increasingly valuable. This post examines four key areas:
- Creative Commons and knowledge-sharing
- Free image resources
- AI art examples and capabilities
- Copyright implications and background on diffusion models
- Relevant tools and resources
Creative Commons¶
The featured "Woman's face and graduated background" image comes from Openverse, a platform where creators worldwide share content under Creative Commons licenses. The example uses CC BY licensing, which permits "distribute, remix, adapt, and build upon the material in any medium or format, so long as attribution is given to the creator" and allows commercial use.
Six Creative Commons license types exist, detailed in the resources section below.
Free Image Resources¶
Before Creative Commons emerged, stock photography was monopolized by a few expensive platforms. The internet democratized image access. Pexels and Pixabay offer authentic, less formulaic images compared to traditional stock sites.
Expert Tip: Use Photoshop's RGB sliders on downloaded images. Adjusting the green slider to where the color graph begins improves most images.
AI Art Examples¶
"Text to image" generators convert written prompts into visual content. Testing the prompt "woman wearing floppy hat and sunglasses" across NightCafe, DALL-E, DreamStudio, and StarryAI produced results in approximately ten seconds.
The same RGB adjustment guidance applies to AI-generated images for quality improvement.
Copyright Implications¶
Diffusion models function as generative systems. According to Ryan O'Connor on assemblyai, they work "by destroying training data through the successive addition of Gaussian noise and then learning to recover the data by reversing this noising process."
Copyright Status: The U.S. Copyright Office has consistently ruled that AI-generated images lack human authorship necessary for copyright protection. Artist Stephen Thaler challenged this in 2022, requesting reconsideration of the Office's 2019 ruling denying copyright for his "Creativity Machine" creation.
Risks: As James Vincent noted on The Verge, releasing unrestricted AI image generators represents "essentially uncharted territory" with unclear consequences regarding potential misuse.
Conclusion¶
AI will shape web imagery's future—not through replacing artistic talent, but by empowering artists to leverage these tools for continued creative innovation and inspiration.
Resources¶
AI Art Tools:
- NightCafe
- DALL-E 2
- DreamStudio
- StarryAI
Free Image Platforms:
- Pexels
- Pixabay
- Unsplash
- Burst
Reference Materials:
- Creative Commons Openverse
- Creative Commons License types
- Alex Katz (artist website)
- Diffusion models explained (assemblyai)
- Google Imagen
- "Anyone can use this AI art generator – that's the risk" (The Verge)