Glaze Protection from AI
TLDRThe panel discussion, moderated by Mia Araho, addresses the concerns of artists regarding generative AI and its impact on their work. Panelists include John Lamb, who initiated the 'create don't scrape' movement; Ben Jiao, a computer science professor working on AI security; Steven Zapata, a designer and illustrator advocating for artists' rights; and Carla Ortiz, a concept artist involved in a lawsuit against AI companies. They discuss the potential misuse of AI in art, the importance of protecting artists' work, and the development of tools like Glaze and Nightshade to combat unauthorized use of their art by AI models. The conversation also touches on the need for collective action among artists and the potential consequences of AI on the creative industry.
Takeaways
- 🎨 The 'Create Don't Scrape' panel addresses the concerns of artists regarding generative AI technologies like MidJourney, DALL-E, and Stable Diffusion, which they feel are exploiting their work without consent.
- 🔍 Panelist John Lamb highlights the impact of generative AI on artists' work, emphasizing the emotional and professional damage caused by AI models using artists' creations without recognition or compensation.
- 🛡️ Ben Jiao, a professor of computer science, discusses his transition from unawareness to active involvement in protecting artists against AI misuse, highlighting his developments in AI security systems.
- 👩🎨 Carla Ortiz shares her journey from a concept artist to becoming a plaintiff in a class action lawsuit against companies using generative AI unethically, reflecting her advocacy for artists' rights.
- 🤖 Steven Zapata explains the basics of how generative AI works, focusing on the issues with how these models are trained using large datasets scraped from the internet, often including copyrighted material.
- 🚫 The panel discusses the unique threat of generative AI to creative professions, contrasting it with other technological advancements that have historically been tools aiding in artistic creation, not replacing artists.
- 💡 Ben Jiao introduces 'Glaze', a tool developed to protect artists' works from being misused by AI by subtly altering images in ways that disrupt AI training processes without affecting human appreciation.
- 🔮 Nightshade is revealed as a more aggressive approach to protect artistic works, designed to 'poison' AI training data, causing models to learn incorrect information if they use copyrighted art without permission.
- 🌐 The panel highlights the importance of community and collective action in combating the unethical use of AI, with tools like Glaze being made freely available to empower artists to protect their works.
- 📢 The discussion concludes with a call to action for artists to stay informed and proactive in using available tools and legal avenues to safeguard their intellectual property against misuse by generative AI.
Q & A
What is the primary concern of the panel discussion titled 'Create Don't Scrape: Glaze Protection Against AI'?
-The primary concern of the panel discussion is to address the threats and challenges posed by generative AI technologies like MidJourney, DALL-E, and Stable Diffusion to artists and their work. The panel aims to advocate for protections for artists and other industries affected by these technologies.
Who is John Lamb and what issue did he raise regarding AI and art?
-John Lamb is described as a 'perpetual poster on Twitter' who became actively involved in the conversation about generative AI after witnessing its impact on artists. He raised concerns about AI technologies misusing the artworks of deceased artists, such as Kim Jung Gi, and the lack of compensation or recognition for artists whose works are used by AI models.
What is the purpose of the hashtag #createdontscrape?
-The hashtag #createdontscrape was created to promote the idea of generating original content rather than scraping existing artworks from the internet to train AI models. It is part of a broader advocacy to protect artists' rights and prevent unauthorized use of their work.
What significant action did Carla Ortiz take in response to the advent of generative AI?
-Carla Ortiz, a fine artist and illustrator, became one of the first plaintiffs in a class-action lawsuit against companies like Stable Diffusion and DeviantArt. She took legal action to challenge the use of artists' work without permission or compensation, advocating for the rights and protections of artists.
What is Glaze, as mentioned by Ben Jiao?
-Glaze is a project developed by Ben Jiao and his team at the University of Chicago. It aims to create protective measures for artists' works against unauthorized AI use. Glaze involves modifying images in such a way that they disrupt AI's ability to accurately learn and replicate the artists' styles.
How does Nightshade differ from Glaze?
-While Glaze is designed to protect individual artworks from being mimicked accurately by AI, Nightshade is a more aggressive approach that aims to 'poison' AI models by training them with deliberately incorrect data. This would potentially disrupt the AI's learning process, making it unreliable for future use.
What concerns were raised about the impact of AI on the job market?
-The panel discussed concerns that AI technology might replace human jobs, particularly in creative fields. There is a fear that AI could increase productivity to a point where fewer artists are needed, thus leading to job losses and reduced human creativity in the art industry.
Why did Carla Ortiz testify in front of Congress?
-Carla Ortiz testified in front of Congress to discuss the impacts of generative AI on the art industry and advocate for legislative action to protect artists' rights and regulate the use of AI in using existing artworks without proper authorization.
What is the significance of artists participating in the development of AI defensive tools?
-Artists' participation is crucial in developing AI defensive tools like Glaze because they bring a unique perspective and understanding of creative processes that can help in designing effective protections against AI misuse of artworks.
How can artists and the public contribute to or support the initiatives against generative AI's misuse of art?
-Artists and the public can support initiatives against AI misuse by raising awareness, participating in legal actions, using tools like Glaze to protect their artworks, and supporting organizations like the Concept Art Association, which advocates for artists' rights.
Outlines
🎤 Introduction to the Panel on AI and Art
The panel, moderated by Mia Araho, discusses concerns about generative AI's impact on artists. Panelists include John Lamb, an advocate for artists' rights online, Professor Ben Jiao from the University of Chicago focusing on AI security, Steven Zapata, a designer and illustrator concerned with artists' well-being, and Carla Ortiz, a concept artist affected by AI's impact. They share personal stories and their paths to advocacy, aiming to create a safe space for artists and raise awareness about AI's effects on creativity and copyright.
🤖 Understanding Generative AI and Its Impact on Artists
Steven Zapata explains generative AI tools, how they work, and the process of training these models using vast datasets from the internet, which include artists' works without consent. The panel discusses the unique threat AI poses compared to other tools, as it consumes and mimics the creative work of artists, potentially cutting them off from potential clients and devaluing their originality.
🖼️ The Role of AI in Art and the Response from the Artist Community
The panelists elaborate on the misuse of AI, its consequences for the art industry, and the emotional toll it takes on artists. They discuss the inception of the 'Create Don't Scrape' movement and the website crap.com as resources to fight against AI's misuse. The conversation also touches on the legal actions taken against companies leveraging AI to exploit artists' work.
🛡️ The Development of GLAZE: A Protective Tool for Artists
Ben Jiao details the creation of GLAZE, a tool designed to protect artists from AI misuse by altering their work in a way that is indiscernible to humans but disrupts AI training. The development of GLAZE is a result of collaboration between artists and the University of Chicago team, aiming to provide a solution against AI's predatory practices on creative work.
🔍 How GLAZE Works and Its Significance for the Artistic Community
The panel delves into the technical aspects of how GLAZE functions, making small changes to images that confuse AI systems while remaining visually unchanged to humans. They discuss the importance of GLAZE as a protective measure and its role in the broader fight against unauthorized use of artists' work by AI models.
🌟 The Future of Artistic Expression in the Wake of AI
The discussion highlights the ongoing efforts to support artists in the face of AI's challenges, including the development of new tools and legal strategies. The panelists express optimism about the artist community's resilience and its ability to adapt and overcome the threats posed by generative AI.
📈 The Economic and Ethical Implications of AI in Creative Industries
The panelists address the potential economic fallout and ethical considerations of AI in creative fields. They predict job displacement and the concentration of profits among higher management, while workers face increased workloads and unemployment. The discussion also covers the importance of unionization and collective action in protecting creative professionals from the adverse effects of AI.
🏛️ Legal and Regulatory Responses to AI Infiltration in Art
The panel touches on legal and regulatory actions that could influence the future of AI in art. They mention upcoming FTC decisions and court cases that may affect the legality and commercial use of existing AI models, emphasizing the importance of the current moment for advocacy and action.
Mindmap
Keywords
💡Generative AI
💡Mid Journey Dolly
💡Stable Diffusion
💡Create Don't Scrape
💡AI Advocacy
💡Glaze Protection
💡Litigation
💡Opting Out
💡Tech Bro
💡Class Action Suit
💡Concept Art Association
Highlights
The panel discusses the impact of generative AI on artists and their work, focusing on concerns about AI technologies like Dolly and stable diffusion.
John Lamb shares his personal journey in advocating for artists' rights after witnessing the effects of generative AI on the art community.
Ben Jiao, a computer science professor, explains his team's development of 'Glaze', a tool designed to protect artists' work from being misused by AI.
Steven Zapata, a designer and illustrator, talks about his advocacy work and the importance of educating artists about the threats posed by generative AI.
Carla Ortiz, a fine artist, discusses her experience as a plaintiff in a class action suit against companies using generative AI and her role in advocating for artists' rights.
The panelists emphasize the unique challenges generative AI poses to artists, contrasting it with other technological advancements in art.
Glaze is introduced as a method to alter images in a way that confuses AI models, preventing them from learning and replicating the original art style.
The concept of 'Nightshade' is teased as a more aggressive approach to combating AI misuse of art, potentially disrupting the training of AI models.
Carla Ortiz describes her emotional experience creating art for the Glaze project, highlighting the therapeutic aspect of being involved in protective measures.
The panelists address the importance of community involvement and the significant response from artists participating in user studies for Glaze.
Ben Jiao explains the technical aspects of how Glaze works, leveraging the fundamental differences in perception between AI and human vision.
The discussion touches on the legal and ethical considerations of AI in art, including the potential for litigation and regulatory actions.
The panelists stress the need for ongoing efforts to protect artists from the misuse of their work by AI, and the role of tools like Glaze and Nightshade in this fight.
The audience is encouraged to use Glaze for their art to contribute to the defense against AI models that may exploit their work.
Ben Jiao addresses the future development of Glaze, acknowledging ongoing efforts to improve its effectiveness against AI models.
The panel explores the broader implications of AI on the job market, suggesting that increased productivity may lead to job losses rather than improved work-life balance.
The discussion concludes with a call to action for artists to glaze their work and a Q&A session where the panelists answer questions from the audience.