Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

progressoid

progressoid's Journal
progressoid's Journal
December 14, 2022

The End of Art: An Argument Against Image AIs




Follow The Concept Art Association for further news on organizing against AIs, and check out their recent AI town hall video: https://www.youtube.com/c/ConceptArtAssociation

Equity is fighting against AIs replacing many kinds of artists, their efforts are relevant to all creatives and they have good resources here: https://www.equity.org.uk/getting-involved/campaigns/stop-ai-stealing-the-show/

Know more organizations going against the AIs? Send me links at: stevensketches (at) gmail.com

A transcript of this video is available here: https://docs.google.com/document/d/128yey0VfYhM9eUdvkvCpk5zvvoIkqXfI4hEPAYeJCHU/edit

A site to investigate just one of the LAION data sets to see if you or your art are in there. A note that this site offers to sign you up for a future opt-out product- I have no idea what the nature of that product will be and would urge caution: https://haveibeentrained.com/

Open AI explaining how they invented their legal structure because nothing else worked for them: https://openai.com/blog/openai-lp/

Stable Diffusion release info with no mention of “artists”: https://stability.ai/blog/stable-diffusion-announcement

Imagen release info that explains how the data set it is trained on is “uncurated” and contains “a wide range of inappropriate content including pornographic imagery, racist slurs, and harmful social stereotypes.” This is another page you should do a ctrl-f search for “artist” on: https://imagen.research.google/

Dance Diffusion explanation: https://wandb.ai/wandb_gen/audio/reports/Harmonai-s-Dance-Diffusion-Open-Source-AI-Audio-Generation-Tool-For-Music-Producers--VmlldzoyNjkwOTM1

Example of private medical imagery being used to train the AI: https://www.vice.com/en/article/3ad58k/ai-is-probably-using-your-images-and-its-not-easy-to-opt-out

A great article summarizing the data laundering techniques of AI companies: https://waxy.org/2022/09/ai-data-laundering-how-academic-and-nonprofit-researchers-shield-tech-companies-from-accountability/

California Congresswoman Anna G. Eshoo writes OSTP urging investigation into unsafe AI release models, namely Stable Diffusion: https://eshoo.house.gov/media/press-releases/eshoo-urges-nsa-ostp-address-unsafe-ai-practices

The best non-technical explanation I’ve seen on how diffusion models work: https://www.tiktok.com/t/ZTRHrUyDM/

A good semi-technical explanation of how diffusion models like Stable Diffusion work:
&t=0s

I did the drawing for this video, “Prey For A Spiritual Creature” on 18”x24” Strathmore bristol paper. From imagination for the sheer love of drawing!

Special thanks to my wife for her help editing the script, to my Patrons for their support, to my friends on my Discord for links, expert info, and first reactions, and to those who get the course, of course.

Profile Information

Gender: Do not display
Member since: Wed Nov 10, 2004, 06:28 PM
Number of posts: 49,961
Latest Discussions»progressoid's Journal