Understanding artificial intelligence is a crucial skill for both artists and the public in the digital age
Artificial Intelligence (AI) is a transformative force reshaping numerous sectors, including arts, creative industries, and many office jobs. As AI becomes increasingly pervasive, it is crucial for everyone - artists, creative professionals, and the general public - to grasp its fundamentals, implications, and nuances. By understanding how these systems are built, evaluated, and deployed, we will be able to use them more effectively and responsibly.
During the Salzburg Global program on "Creating Futures: Art and AI for Tomorrow's Narratives", AI literacy was identified as one of the key areas for future development. Fellows of the program, including activist Micaela Mantegna and artists Phaan Howng and Eddie Wong, contributed to the discussions and the creation of the group’s AI Literacy Curriculum.
Phaan Howng is an American-born Taiwanese artist making paintings, immersive installations, and performances on the theme of Earth’s landscape in a post-human future. From her perspective, “AI is ableist against those who are computer or technology illiterate...and don't care”. For Phaan, there is a need to acknowledge that AI is driven by humans: “The ‘information and knowledge’ supplied to AI platforms is populated with text, images, data, as uploaded by humans with access to internet, software, social media and media platforms. To be ‘AI literate’ is to equip individuals with the knowledge, skills, and ethical framework to recognize this situation, and that those who use AI should proceed with mindfulness–such as checking facts, biases, and sources.”
Phaan emphasized the necessity of prioritizing ethics: “Given the significant human influence on AI, and vice versa AI’s influence on humans, it is important to place a strong emphasis on ethics. We need to think critically about the backend of AI development, considering factors such as who is designing the platforms, who owns the datasets, who develops the business models, and who has the right to train and control AI models.” To achieve this, Phaan advocates for a public service announcement (PSA): “It's imperative that a huge PSA about AI has to be made aware to the general public, especially now that big tech companies are pushing it on us by integrating it into their software, apps, and search engines without our understanding of AI and our consent. I think this PSA has to come from the government and not from the tech companies themselves and be taught in schools.”
Meanwhile, Eddie Wong, a Malaysian artist and filmmaker, commented on the specific requirements for artists: "AI literacy for artists isn't just about mastering new tools; it involves a multifaceted understanding of the technical, philosophical, artistic, and ethical dimensions of machine learning. ML models are not static tools but dynamic systems that create emergent outputs, constantly evolving based on the data they process, often leading to unexpected results. Dataset provenance is a crucial factor, including who designs the platforms, who owns the datasets, who develops the business models, and who has the right to train and control AI models. To achieve AI literacy, artists should grasp the technical aspects of AI, such as machine learning and neural networks, and critically engage with its ethical and societal implications.”
Eddie’s perspective emphasizes the importance of artists engaging with AI on multiple levels. They must understand not just how to use AI tools, but also the underlying mechanics, data provenance, and ethical considerations of these technologies. This comprehensive approach will allow artists to leverage AI creatively while highlighting its limitations and being mindful of its broader impacts.
As for the impact of AI on artists themselves, there have been a number of projects over the past few months that deal with companies training AI systems on artists’ works, typically without their consent. Research projects such as Nightshade allow artists to “poison” their data so that it causes AI models to behave in unpredictable ways, while initiatives by artists such as Holly Herndon’s Spawning AI and Ed Newton-Rex’s Fairly Trained offer solutions for both artists and AI developers to declare and respect data preferences.
However, this is not the only type of intellectual labor that powers the AI systems of today. In her article “ARTificial: Why Copyright Is Not the Right Policy Tool to Deal with Generative AI” for the Yale Law Journal, Micaela Mantegna, a video games lawyer and activist known as the “Abogamer”, looks at the types of labor involved in the AI development process and the different forms of recognition they receive. She states that “OpenAI has acknowledged that models would be irrelevant or ineffective without training on copyrighted materials. If so, the same is true about ‘ghost work’ - the contributions of data workers globally are indispensable to fine-tune and sanitize AI models. Recognizing one form of intellectual labor within the copyright framework while neglecting to acknowledge equally relevant contributions of other types of labor results in an unfair distinction that privileges one type of worker over another.”
Micaela highlighted the need for fair treatment of all contributors to AI systems: "Intellectual labor is particularly evident in the context of the labor inputs used in the development of generative AI (GAI) models, where both creative works and other types of more pragmatic intellectual labor are intertwined in the training process. Privileging creative intellectual labor over data-classification labor creates an unfair hierarchy within the AI industry and further stigmatizes ghost workers. Unlike copyright, policy interventions would allow regulators to design a uniform approach that recognizes and compensates all contributors alike.”
As emphasized by Phaan, Eddie, and Micaela, AI literacy is a crucial element for the future advancement of our society and a key pillar in ensuring that AI systems are developed in a way that takes everyone’s needs and contributions into account. AI literacy is not just about mastering technical tools, but also about understanding the ethical, societal, and labor implications of AI. By incorporating these insights, we can build a more comprehensive and equitable approach to AI literacy, benefiting individuals and society as a whole.
The Salzburg Global Fellows featured in this article convened alongside around 50 other artists, technologists, futurists, curators, and activists for Salzburg Global's annual Culture, Arts and Society program in May 2024. "Creating Futures: Art and AI for Tomorrow's Narratives" explored the emergent possibilities at the intersection of creative expression, technology, and artificial intelligence.
This article featured in our Shorthand story, which includes more coverage from the program "Creating Futures: Art and AI for Tomorrow's Narratives".
