PREFACE
In spring 2023, when the world was abuzz over generative AI, I attended an art exhibition in Midtown New York. Wanting to delve deeper into the stories behind the art, I was drawn to the exhibition catalog. As I thumbed through its pages, absorbing the rich history and context of each piece, an unexpected note on the copyright page made my heart skip a beat:
“The research for this catalog has been generated through ChatGPT and edited by…”
I blinked, reading the statement again. Here it was, ChatGPT, not in some abstract discussion but right in front of me, woven into the fabric of the art world. The reality of AI’s encroachment into traditional roles became tangible.
As a trustee for YoungArts, I passionately believe in the organization’s vision that artists are vital to our humanity. Seeing these technological advancements, I felt—and continue to feel—that artists face an existential threat. We may not be able to halt the progress of technology, but we can equip artists with the tools to fight back.
That’s why I created this series of guides: to demystify generative AI for artists and provide a roadmap for exploration.
I admit, even though I have made a career working with emerging technologies, AI scares me. It may evoke fear, skepticism, or discouragement in you too. But through knowledge comes opportunity, and through opportunity comes power.
Technology can enable great achievements and push humanity to new heights. Take, for instance, the visionary architect Frank Gehry. He once remarked that without the aid of design software, many of his groundbreaking architectural masterpieces would have remained mere figments of his imagination. In the same vein, while AI might seem daunting now, it also holds great promise.
My hope is that these guides help each of us find our own way to shift the threat of AI into an opportunity for something beneficial and transformative. Humanity deserves it.
Natalie Diggins
Creator, The Art Practitioner’s Guide to Generative AI series
Table Of Contents
The Speed of Innovation: Keeping Pace with AI in the Arts
In the fast-changing world of generative AI, the landscape shifts dramatically, even from day to day. Since these guides were first created, there has been significant evolution. New AI companies have emerged, and established ones have evolved so much that they’re almost unrecognizable. Even referring to these recent companies as ‘established’ underscores the speed of these changes.
To keep pace, my blog discusses the impact of the latest generative AI features and functionalities on artists. For instance, I cover ChatGPT’s voice and image input and output capabilities in this post.
Throughout this guide, I use the terms ‘Generative AI’ and ‘ChatGPT’ interchangeably for simplicity and ease of understanding, though this should not be taken as an endorsement of any particular product. It’s important to note that other AI generators, like Claude, Bard, Midjourney, Stable Diffusion, and others, can also be effectively utilized. The instructions provided in this guide are universally applicable, regardless of the specific generative text input AI tool you choose.
While the examples in this edition focus solely on text-based AI generation, many of the techniques shared in this guide are also applicable to text-to-visual applications. Future editions of the guide may incorporate specific examples of these applications.
Creator's Statement
In the development and refinement of writing the Art Practitioner’s Guide to Generative AI series, I have incorporated the use of OpenAI’s AI tool based on GPT-3.5. This technology played a significant role in enhancing the guides, including narrative development, idea expansion, language enrichment, and proofreading. The voice is distinctively my own.
Rightly, there is considerable concern and debate surrounding the use of material to train ChatGPT and other similar technologies, in addition to the risk of bias. I leave it to the artists, legislators, government officials, and courts to establish the framework governing and compensating such use and attribution, and I will incorporate and follow any rules, protections, standards, or best practices that result.
I specifically chose to use what was at the time OpenAI’s free and text-only tool to closely mimic the experience any artist with access to the Internet could have without the burden of cost.
Introduction: Bridging the Gap between Art and Generative AI
Today’s headlines are dominated by generative artificial intelligence (AI). Many people see this new technology as a cause for fear and uncertainty. But it should also be seen as an opportunity, especially for creators. The goal of this practical guide is to bridge the gap between generative AI technology and the artistry of dance.
THE HOW
This guide is, above all, intended to be practical. In an age awash with theoretical insights, we understand the pressing need for actionable, hands-on knowledge. With the surge of generative AI in the world, artists need more than an overview—they need a roadmap. That’s why we’re not just discussing the ‘what’ and ‘why’ of generative AI—we’re delving into the ‘how.’ We want to equip you with the tools and knowledge to harness AI’s potential.
THE WHY
But why we’re creating this guide is still incredibly important. At the heart of this guide lies our commitment to empowering artists. By introducing you to the potential of generative AI, we’re giving you another tool in your artistic endeavors.
This guide aims to inspire innovation. We demystify generative AI and spotlight real-world artistic examples. AI challenges artists to move beyond convention, to experiment, and envision their craft in groundbreaking ways. Yet, innovation shouldn’t occur in isolation.
Hence, our focus on fostering collaboration. By integrating feedback and collective insights from the artistic community, we’re creating more than a guide; we’re crafting a platform for shared learning and mutual growth.
We also believe in democratizing access to this knowledge. Art shouldn’t be restricted by technological barriers. By distilling complex concepts into approachable insights, we’re striving to level the playing field, ensuring artists from diverse backgrounds can utilize AI.
The world of art and technology is fluid, and we embrace that. Our commitment to stimulating continuous growth means this guide remains an evolving entity. As we garner feedback and as the landscape shifts, we will iteratively refine our content, helping ensure that artists are always at the forefront of innovation.
THE GOLDILOCKS APPROACH
We initially aimed to create a guide that would cater to artists at every stage of their career, but given the rapid pace of technological evolution, in the time it would take to write the comprehensive guide we envisioned, it would likely be outdated by its release.
As a result, we’ve opted for a broader ‘Goldilocks approach.’ Some artists might find our examples too basic, while others might deem them too complex. But we hope that for many, they’ll be just right—and ideally, every artist will find something to enhance their practice.
Our measure of success for this project is simple: if you discover just one piece of valuable information—which could even be that generative AI isn’t for you—we’ve achieved our goal.
Now, let’s continue our exploration, enriching our creative process and reshaping our engagement with the world of dance.
Your Brilliant and Supportive but Unreliable Collaborator
Generative AI is your eager collaborator, albeit an imperfect one. It’s like that endearing friend who loves telling stories but sometimes gets a little too creative with the details. Be sure to bring your fact-checking skills to the conversation.
Think of it as an unblinking, nonjudgmental sounding board, always ready to help you solve a problem at 3 a.m. It can take in your wildest idea, juggle it around, and toss it back with a fresh, unexpected spin.
It is the most accommodating coworker you’ll ever find, ready to dive into virtually any topic at a moment’s notice. Just remember to take its words with a grain of salt, and perhaps a dash of humor.
AI In the Arts - A Tool for Enhancement, not Replacement
AI simply confirms everything we have thought, but beauty in design is some- thing we discover; it emerges from corners you never thought it would.
Reinier de Graaf Partner,
Office of Metropolitan Architecture
While AI possesses extensive knowledge, it doesn’t understand the essence of beauty. AI excels in unveiling the hidden and unpredictable elements of design and artistic expression, but it lacks the human ability to translate that into works full of emotion and impact.
The real promise of AI isn’t about rendering artists obsolete but about magnifying their creative capacity. By incorporating this tool into your repertoire, you can navigate unexplored artistic landscapes and tap into new forms of your dance practice.
This shift can be likened to the evolution of online dance classes. For example, during the Covid-19 pandemic, platforms like Zoom and various fitness apps allowed dance classes to continue in a virtual format. While initially there were fears that this could mean the end of in person dance classes and jobs for dance instructors, what has actually happened is that it’s provided an additional resource for dance students. Technology makes it possible for them to train with teachers they might not have had access to before due to geographical location and provides an extra source of income for dance teachers.
Just as virtual dance platforms unlocked new horizons, generative AI holds the key to a treasure trove of untapped potential. The DANCE Practitioner’s Guide to Generative AI can help you delve into this field.
This hands-on guide is a resource for dance practitioners of all types and abilities, offering insights, guidance, and practical tips to enhance your practice using generative AI. No advanced computer skills are necessary. If you’re comfortable with searching the Internet with Google, you’ll find ChatGPT easy to use.
With the help of ChatGPT, you can conduct early investigative research, generate new movement ideas, get suggested exercises for performance enhancement, receive music recommendations, determine the perfect costume for your performance, and ask questions to expand your understanding of dance techniques, history, and styles, plus much more. Embrace the joy of movement, push your boundaries, and let this guide be your companion on your dance journey.
Let’s get started!
How to Set Up a Free ChatGPT Account
You can interact directly with AI through a tool called ChatGPT. Today, ChatGPT-3.5 is free after creating an account with your email address. For those seeking a more advanced experience, a premium version, ChatGPT 4, is available for a subscription fee. However, while this service is readily accessible in the United States, its availability is limited in certain countries.
To get started, visit the ChatGPT login page and create your account: https://chat.openai.com/auth/login
Have you set up your ChatGPT account? Great. Now let’s explore how you can use this technology in your dance practice.
Interacting with ChatGPT
ChatGPT operates primarily on text inputs, so it’s just like having a conversation with another person, but in this case, it’s a machine.
- Once you’ve logged into your account, find and click on the chat interface.
- You’ll see a blank chat box or a welcoming message. To start a conversation, you just type your message into the input field. This can be a question, a command, or just a general statement.
- Chat is designed to respond to your queries or statements. For example, you could ask it “When was George Balanchine born?” or say “Find me a poem about dance.” It’s capable of answering questions, providing information, generating text like stories or poems, and even engaging in casual conversation.
- After you’ve typed in your message, hit enter or click the “Send” button. Chat will process your message and generate a response, which will appear in the chat box. Response time usually takes just a few seconds.
- You can then continue the conversation based on the AI’s response. Remember, AI is designed to maintain a consistent chat flow, so it will always relate its responses to the ongoing context of the conversation.
- You can end the conversation anytime you want by simply not entering any more text into the chat box.
Technology News You Can Use in your Dance Practice
The Ethics of AI: Walking the Walk
When considering the use of generative AI, whether in the arts or any other field, there are a variety of ethical issues to consider. These issues require meticulous examination and resolution. For example, there may be concerns over bias and fairness, which question the impartiality of AI. There may also be concerns regarding originality and plagiarism, raising doubts about the authenticity of AI-created works. There’s the issue of manipulation and misinformation, where the possibility of AI being used to distort reality or mislead people is a significant concern. Finally, transparency and disclosure are pivotal, necessitating clarity about the use of AI.
While ChatGPT’s response to the issue isn’t comprehensive, or necessarily correct or right for you individually, it can serve as a jumping-off point for further reflection and discussion.
[Update: These suggestions were drafted when generative AI became available to the general public. Since then, text citation practice has evolved. MLA Style, APA Style, Chicago Style and many academic standards have been updated to address textual generative AI contributions.
Outside of academia, practicing consensus has evolved that generative AI is a tool, similar to Microsoft Word’s Editor feature, Adobe Photoshop, and spellchecking software. Similar to how these tools aren’t cited, generally neither is generative AI. That said, while generative AI may be used as a tool, the nature of its contributions to content creation might warrant different considerations compared to more straightforward productivity software.
The bottom line is that there is no right answer. Citation is a work in progress, and you should use your own judgment based on content and context.
The critical and complex issues of creator permission, attribution, and compensation remain unsolved and appropriately controversial.]
PRO TIP: While AI can be a useful tool, it isn’t a substitute for meticulous human review and in-depth follow-up. AI can produce errors and inconsistencies. Especially when delving into historically marginalized topics, cultural representations, or issues of authenticity, complement AI insights with trusted external sources. If you suspect faulty information, a first but not conclusive step is to ask the tool the same question again. Revisiting the question helps in cross-checking the consistency of the information provided. Then, query external sources. Exercise extreme caution and diligence on sensitive or nuanced topics.
Early Investigative Research
ChatGPT can be an invaluable tool dancers can use to explore early investigative research as part of the dance process. Discover historical figures who have blended specific aesthetics or combined various disciplines in the past, examine themes and concepts, and learn how other dancers have overcome specific artistic challenges.
To use ChatGPT for generating early investigative research ideas, follow these steps:
- Set the context: Begin by explaining the general area of interest or the specific themes and concepts the dancer is curious about. For example, “I’m interested in researching dance artists who have combined visual arts and dance in their performances.”
- Specify the aesthetics: Describe the specific aesthetics or artistic elements the dancer wants to explore. This could include movements, visual design, music integration, or any other aspects that intrigue them.
- Ask for examples: Request specific examples of historical figures or artists who have explored similar themes or aesthetics. For instance, “Can you suggest any dancers from the early 20th century who experimented with incorporating elements of nature into their choreography?”
- Seek historical context: Inquire about the historical context and significance of the artists or movements that ChatGPT suggests. For example, “Could you provide some background information about the artistic movement that influenced this particular dance artist?”
- Explore interdisciplinary collaborations: Ask ChatGPT to suggest artists known for their cross-disciplinary collaborations or works that involve other art forms, such as music, theater, or visual arts. For instance, “Are there any choreographers who have collaborated with painters to create dance performances?”
- Refine the investigation: Engage in a dialogue with ChatGPT to further refine the research focus, ask follow-up questions, and explore related topics that arise from the initial responses.
PRO TIP: At the beginning of a conversation with ChatGPT, you can utilize something called a system message. A system message works as an initial instruction to provide context for the AI model. For example, if you’re a dance historian and want to discuss the history of flamenco, you could start the conversation by saying, “As a dance historian with an interest in flamenco, I want to discuss the influence of Romani culture on flamenco’s evolution.” This message helps the AI understand your background and the context of the discussion, allowing it to tailor its responses accordingly.
Movement Ideas
ChatGPT can help you brainstorm and generate new movement ideas. Describe the style, mood, and inspiration you’re looking for, and ChatGPT can offer suggestions and inspiration.
To use ChatGPT for generating movement ideas, follow these steps:
- Describe the style: Begin by explaining the specific style of dance you’re interested in, such as contemporary, hip-hop, ballet, jazz, or any other genre. Provide details about the movement qualities, tempo, dynamics, and any other characteristics you want to incorporate.
- Set the mood: Describe the mood or emotion you want to convey through your movement. For example, you might want something energetic and upbeat, graceful and lyrical, or emotional and introspective. Mention any specific themes or ideas you have in mind.
- Provide inspiration: If you have any particular sources of inspiration, such as a piece of music, a painting, a poem, or a story, mention them to ChatGPT. Explain how these influences can be integrated into the movement or serve as a starting point for the creative process.
- Collaborate and iterate: As ChatGPT generates suggestions, you can engage in a conversation, ask follow-up questions, and provide feedback. Think of this as a conversation you’d have with a colleague or mentor. This interactive process can help refine and narrow down the ideas until you find something that resonates with your vision.
PRO TIP: If ChatGPT doesn’t seem to understand your question, rephrase it. When choreographing a dance, not every step you first plan will fit the music or flow naturally into the next. You might need to revise and try different steps. Similarly, with ChatGPT, if you ask, “How to perform a pirouette?” and it doesn’t give the detail you want, you could try, “What are the steps to execute a pirouette?” or “What techniques are important for a successful pirouette?” Just as you refine your dance moves to achieve the perfect routine, sometimes adjusting your question, often again and again, can lead to a more useful answer.
Performance Enhancement
ChatGPT can recommend drills and exercises to help improve your dance performance. It can propose strengthening exercises, flexibility routines, or targeted drills to develop muscle memory, coordination, and control. ChatGPT can also suggest specific warm-up exercises to prepare your body for the technical demands of the performance.
To use ChatGPT for ideas on drills and exercises, follow these steps:
- Specify your dance style and areas of focus: Describe your specific dance style (e.g., ballet, hip-hop, contemporary) and the aspects of your performance you want to enhance. This could include technique, flexibility, strength, musicality, or any other areas you wish to target.
- Request drills and exercises: Ask ChatGPT for suggestions on drills and exercises that can help you improve in the areas you mentioned. Be specific about what you’re looking to work on and the level of difficulty you desire.
- Seek explanations and examples: Engage in a conversation with ChatGPT to request explanations and examples of the suggested drills and exercises. Ask for clarification on how to perform them correctly and effectively.
- Consider safety and personal limitations: While ChatGPT can offer suggestions, it’s important to prioritize your safety and listen to your body. Adapt the suggested drills and exercises to your skill level and physical abilities. If you have any pre-existing conditions or concerns, consult a qualified dance instructor or healthcare professional before attempting new exercises.
- Experiment and track progress: Incorporate the suggested drills and exercises into your dance training routine. Stay consistent, track your progress, and observe how these new additions contribute to your performance improvement. Adapt and modify them as needed based on your personal feedback and observations.
PRO TIP: Know that ChatGPT won’t always give the same answer to a question each time it’s asked. When performing a dance, you may have the same piece of music and the same choreography, but every time you perform, you bring something different to the stage — a different emotion; a slight variation in pace or intensity. Similarly, when you ask ChatGPT a question, even though the information base is the same, it might provide a different “interpretation” or answer each time.
Music Selection
ChatGPT can recommend music that aligns with your dance style or concept. You can provide details about your performance, and ChatGPT can suggest songs that might suit your needs.
To use ChatGPT for music selection ideas, follow these steps:
- Specify your dance style or concept: Explain the type of dance you’ll be performing or the specific concept you have in mind. Describe the overall mood, tempo, and energy level you’re aiming for. For example, you might mention if it’s a contemporary piece, a high-energy hip-hop routine, or a graceful ballet performance.
- Provide details about your performance: Share any specific movements, motifs, or moments that are key to your choreography. This information can help ChatGPT understand the dynamics and timing required in the music selection.
- Mention preferences or restrictions: If you have any preferences for the music genre, artist, or specific songs, let ChatGPT know. Additionally, if there are any restrictions or guidelines, such as music licensing concerns or specific requirements for a competition or event, mention those as well.
- Explore options and refine choices: ChatGPT can generate a list of potential songs or music tracks based on the information you provided. Discuss and explore the suggestions, ask for additional options, or request more details about the recommended tracks. You can also ask for specific genres or artists to be considered.
- Listen and evaluate: Once ChatGPT suggests songs, listen to the tracks it recommends. Consider how well they match your style, concept, and desired mood. You may need to iterate and provide further feedback to ChatGPT to narrow down the choices until you find the perfect music for your performance.
PRO TIP: Ask ChatGPT progressively deeper questions, much like refining a dance routine with your students. Initially, you might start with a broad question like, “How can I teach contemporary dance techniques?” As you unpack the answer, narrow your focus based on the specifics you’re seeking, similar to how you’d break down complex dance movements for your students. Your next query might be, “What are the key principles of contemporary floorwork?” or “How can I incorporate fluid transitions in contemporary choreography?” By continuing this approach, just as you’d gradually build upon foundational dance steps to create a more intricate sequence, you’ll obtain detailed, layered insights from ChatGPT that can immensely benefit your teaching strategy and, consequently, your students’ learning.
Costumes
ChatGPT can assist you in choosing costumes by providing ideas, suggestions, and considerations to enhance the visual appeal and practicality of your performance.
To use ChatGPT for costume selection ideas, follow these steps:
- Provide context: Begin by explaining relevant details about your dance performance, such as the genre, theme, mood, or specific requirements.
- Specify your preferences: Share any specific colors, styles, or materials you have in mind. You can also mention any particular elements or themes you want to incorporate.
- Describe the desired aesthetic: Describe the dance theme and any insights on styles, colors, textures, and embellishments that may enhance the overall look and feel of your performance. Include requirements on costume accessories, such as footwear or props, that can complement the performance.
- Consider practicality: State any practical considerations you have, such as comfort, ease of movement, and suitability for your performance.
PRO TIP: : Imagine that ChatGPT is like a costume designer for your dance number. If the costume it creates allows you to perform your movement seamlessly and within the theme, ask for more outfits that maintain this balance of form and function. If, however, the costume hinders your dance moves or doesn’t quite match your vision, don’t hesitate to provide feedback so it can tailor the next design to better match your dance.
Securing Dance Roles and Jobs
ChatGPT can help navigate the intricate and highly competitive realm of professional dance opportunities. Whether your aspirations lean towards dancing roles, teaching, choreography positions, or other engagements, ChatGPT can help you craft a roadmap to maximize your chances and succeed in this dynamic field.
To seek advice on securing dance jobs using ChatGPT, follow these steps:
- Be Specific in Your Request: Clearly state what you’re seeking assistance with. For example, “I need advice on preparing for dance auditions” or “Can you provide tips for networking in the dance industry?”
- Provide Relevant Details: Include important information such as your dance background (e.g., dance style, level of experience), the type of dance position you’re interested in (e.g., ballet company, Broadway musical), and any specific challenges or questions you have.
- Ask Clear and Direct Questions: If you have specific questions or need guidance on particular aspects of your dance career, pose them directly. For example, “What are some effective strategies for creating a standout dance portfolio?” or “How can I find auditions for contemporary dance roles?”
- Request Actionable Advice: Ask for practical advice or steps you can take to achieve your goals. For instance, “Please provide tips on improving my dance audition technique” or “What are the key elements to include in a professional dance resume?”
- Consider Any Special Requirements: If there are unique aspects or constraints related to your situation, make sure to mention them. For example, “I have limited access to in-person dance classes due to my location, can you suggest online resources for training?” or “I’m looking to transition from ballet to contemporary dance; what should I focus on during this transition?”
- Consider Context: If your request relates to a previous interaction or specific circumstances, briefly mention this context to provide a background for your question.
PRO TIP: Understand the dual nature of ChatGPT’s knowledge. It holds information on performances, dancers, and pieces up until its last update of April 2023. For developments after April 2023, ChatGPT automatically searches the internet for the most recent updates. So while you may hear that ChatGPT’s training data isn’t current, it does have the ability to access the Internet for current information.
The Art of Asking the Right Questions
By now, you may have picked up on a common thread running through our exercises: there’s an art to asking ChatGPT the right questions to draw out the information you’re seeking. This skill is known as “prompt engineering,” and its application is not limited to text creation—it’s an indispensable tool for interacting with any generative AI technologies, even those that translate text into visuals, such as DALL-E, Midjourney, and Stable Diffusion.
The framing of prompts, much like the phrasing of a question in human interactions, can significantly influence the response. This is reminiscent of the legal concept of “leading the witness,” where the way a question is posed can inadvertently or intentionally guide someone to a particular answer.
For example, asking the leading question, “Considering the controversial and shocking nature of art exhibition A, how has it been received by the public?” versus asking the neutral question, “What are the themes of art exhibition A, and how has the public responded to them?” yields very different responses.
The leading prompt assumes the exhibition is controversial and shocking. Meanwhile, the neutral version first seeks to understand the themes of the exhibit and then gauge the public’s reaction.
In the context of generative AI, a poorly framed or leading prompt may result in biased, incorrect, or undesirable outputs. Conversely, a well-phrased prompt can harness the AI’s capabilities more effectively, generating better, more accurate responses.
Although some companies are currently developing tools to automate this task, mastering prompt engineering gives you a unique advantage when interacting with various AI products, not just ChatGPT. Indeed, for the field of generative AI, being proficient in prompt engineering is like knowing how to read in a library full of books. It’s a key to unlocking a lifetime of learning.
Your Turn
Now, think about your own creative practice. What questions do you have that might benefit from ChatGPT’s input? Remember, unlike with a search engine where you ask a single question and the engine returns an answer, dialoging with ChatGPT is a conversation.
Here are some starter questions to help you begin:
- I am currently recovering from a recent bout of sciatica, which primarily affects my lower back and extends down to my right leg. Given that I am undergoing physical therapy once a week, could you recommend additional exercises or daily practices that might aid in my recovery process and help alleviate the pain? I am particularly interested in exercises that can be performed at home with minimal equipment. Creator’s note: ChatGPT is not a substitute for medical advice. Always consult medical professionals before making any changes to your recovery regimen based on its recommendations.
- I am planning a dance routine for a community event, which I also intend to share on platforms like TikTok. I am particularly interested in using electronic dance music. Could you guide me on how to legally obtain usage rights for these tracks for both live and digital distribution? I am looking to understand both the process and any potential costs involved. Additionally, if you have suggestions for platforms or organizations that facilitate such rights acquisition, it would be greatly appreciated.
- I’m developing a dance choreography project currently suited for intermediate dancers. I aim to broaden its appeal to include individuals with diverse dance skills and abilities, both through live performances and online platforms. Could you recommend strategies to make the choreography more inclusive and tips for effectively using digital platforms to reach a diverse audience?
- I am keen on learning about Okwui Okpokwasili, a renowned artist known for her multidisciplinary approach. Could you provide an overview of her background and notable works? Additionally, I’m interested in understanding her distinctive style of dance choreography. Are there any specific techniques or philosophies that she employs?
- I have a dance performance coming up and I’m quite nervous, primarily about maintaining a confident stage presence and executing my moves correctly. Could you suggest a condensed preparation guide that includes mental readiness and physical warm-up exercises?
These examples are intended merely as a starting point. Dive in. Experiment, iterate, but most importantly, don’t worry—you’ve got this.
How do you use Generative AI?
Your insights could greatly enhance this guide, providing invaluable support to our dance community. Here’s how you can help:
Share your experiences: If you have already explored generative AI in your artistic practice, we would love to hear about it. How has it impacted your creative process? What challenges have you faced? What benefits have you discovered? By sharing your stories, you can inspire and guide fellow artists who are also exploring how generative AI can benefit them.
Contribute to the guide: We welcome your thoughts and suggestions for improvement. Is there anything missing? Are there sections that could be expanded or clarified?
Thank you for joining this journey. Every step we take in exploring the potential of generative AI not only enhances our individual artistic practice but also strengthens our collective dance community. So keep dancing, keep experimenting, and let’s continue this adventure together.
Please share feedback with Natalie Diggins @ natalie@thearts.ai
Also available in The Art Practitioner’s Guide to Generative AI series:
–A Classical Music Practitioner’s Guide to Generative AI
–A Jazz Practitioner’s Guide to Generative AI
Glossary
Artificial Intelligence: Artificial intelligence (AI) is the simulation of the thinking, learning, and problem-solving abilities of humans by computers or other machines.
ChatGPT: ChatGPT is a type of generative AI that’s trained to understand and generate human-like text. By learning from billions of sentences, it can have conversations, answer questions, and even write stories that sound like they were written by a person.
DALL-E: DALL-E is an AI image generator model developed by OpenAI. Like Midjourney and Stable Diffusion, it can create new images in response to text descriptions or even from random inputs. Think of DALL-E as an artist that can draw a new, unique picture based on a story or description you give it.
Generative AI: Generative AI is a type of artificial intelligence that can create new content, such as images, text, or music, by learning patterns from existing data. Generative AI learns from a lot of examples, identifies patterns within them, and then uses those patterns to create new, original content. For instance, after analyzing many pictures of dogs, it can draw a new picture of a dog that doesn’t exist.
Midjourney: Midjourney is an AI image generator model developed by Midjourney, Inc. Like DALL-E and Stable Diffusion, it can create new images in response to text descriptions or even from random inputs. What sets it apart is its proficiency in adapting art styles that can be mixed and matched to create an image.
Prompt Engineering: Prompt engineering is the process of carefully crafting questions or commands (prompts) for AI systems like ChatGPT, with the goal of getting more useful or accurate responses. It’s like learning to ask the AI in the right way to get the best possible answer.
Stable Diffusion: Stable Diffusion is an AI image generator model developed by Stability AI. Like DALL-E and Midjourney, it can create new images in response to text descriptions or even from random inputs. It can start with a blank, messy canvas (which is called “noise”) and gradually refine it into a new image. In AI terms, this “noise” is just a scramble of random colors and shapes. But over time, the AI slowly adjusts this mess, bringing out details and structure, until a recognizable image forms. It’s a bit like watching a blurry photograph gradually come into focus.
System Message: A system message, in the context of a chatbot conversation like with ChatGPT, is a directive or instruction given at the beginning of the conversation that sets the context, role, or behavior of the AI. This allows the user to guide the AI’s responses and tailor the conversation to their specific needs or objectives.
About Us
Natalie Diggins
Natalie Diggins is a technology founder, hands-on executive, investor, and advisor with over two decades of strategic and operational experience in engineering and operations in both public and venture-funded technology enterprises.
Throughout her career, Natalie has been instrumental in scaling nascent technologies and methodologies that are now at the forefront of the tech world, including artificial intelligence, machine learning, robotics process automation, blockchain, cybersecurity, Internet of Things, and cloud technologies.
Natalie serves on the board of trustees of YoungArts: The National Foundation for the Advancement of Artists, and she is passionate about ensuring that artists have a voice in shaping the direction of technology and its implications for society. In line with this belief and her commitment to democratizing technology access, she recently founded TheArts.ai, an initiative designed to bridge the gap between technology and the arts
YoungArts
Established in 1981 by Lin and Ted Arison, YoungArts identifies exceptional young artists, amplifies their potential, and invests in their lifelong creative freedom. YoungArts provides space, funding, mentorship, professional development and community throughout artists’ careers. Entrance into this prestigious organization starts with a highly competitive application for talented artists ages 15–18, or grades 10– 12, that is judged by esteemed discipline-specific panels of artists through a rigorous blind adjudication process.
With Thanks
My heartfelt thanks to the remarkable artists who generously dedicated their time and expertise to review this guide, helping to shape it into a more useful and accurate tool. Special appreciation goes to Lauren Slone, Emily L. Waters, and Mario Alberto Zambrano for their invaluable contributions. Any errors or omissions are solely my own.