Team: 15
School: Cleveland High
Area of Science: Computer science
Interim: Problem definition:
Artificial intelligence art piece, AI art generator is an artificial intelligence that analyzes thousands of images and with human assistance creates art. first developed by Harold Cohen in 1973. AI art has created a stir in the art industry as it grows in popularity. With AI art being accessible with just an online search the average person can make their own art by just describing what they want it to look. This has caused trouble because artists, character designers, and etc fear that they might lose their jobs to AI. To make things worse you can ask the AI to make an art piece in the style of any artist. Most AI generators use art from the web to use as datasets to help it generate images. This has become a problem since artists complain that the AI is copying the images since they look so similar to the original piece. This can cause trouble for someone who wants to make money selling art because now anyone can get an art piece and steal the artist's style. With Ai art becoming more widespread it's hard to tell the difference between human made art and AI generated art. The goal of this project is to create a program that can differentiate between human made art and artificially made art.
Problem to solution:
The program would be run in a jupyter notebook. There will be images of AI created art and human made art and the program will learn from the images and recognize the pattern in them to determine which art is artificial or not. This program will focus on human made art and the AI art generator DALL-E.
Progress to Date:
Presently the program is being created in a Jupyter notebook. I am currently collecting images generated by the AI art generator DALL-E to use in my project. Human made art will also be collected for the project to use for data.
Expected results:
After many runs and trials and errors the program will be able to accurately tell the difference between DALL-E made generated art and human made art. The program could be relied upon if you want to know if you're looking at human made art. This program could help pave the way for more future programs that help you know which art is artificial or not. This program could become more useful in the future when AI art becomes more frequent. In the future this program could be expanded to not just DALL-E generated art but also many other art generators as well.
Citations:
Gu, L., & Li, Y. (2022, July 11). Who made the paintings: Artists or artificial intelligence? the effects of identity on liking and purchase intention. Frontiers. Retrieved January 10, 2023, from https://www.frontiersin.org/articles/10.3389/fpsyg.2022.941163/full
Lawson-Tancred, J. (2022, September 14). Will A.I. usher in the end of human artists? fear not, some say. Artnet News. Retrieved January 10, 2023, from https://news.artnet.com/art-world/ai-generated-art-debate-2175570
Costa, R. (2022, December 16). What does the future of copyright look like with AI art? no one really knows. CultureSlate. Retrieved January 10, 2023, from https://www.cultureslate.com/editorials/what-does-the-future-of-copyright-look-like-with-ai-art-no-one-really-knows
Staff, P. (2022, December 16). AI art is taking over social media, but creators are divided. Passionfruit. Retrieved January 10, 2023, from https://passionfru.it/ai-art-creators-divided-2664/
G., T. (2021, December 4). The role of AI in art creation. Medium. Retrieved January 10, 2023, from https://towardsdatascience.com/the-role-of-ai-in-art-creation-a53dbd562cdb?gi=406125aae91c
Team members: Layla Barela
Sponsoring teacher: Ashli Knoell
Team Members:
Sponsoring Teacher: Ashli Knoell