Cory Yihua Li is a New York-based media artist, Vtubing enthusiast, and CGI specialist exploring AI-driven creativity. His work has been featured in NYC ARTECHOUSE, DEJI Art Museum (with Beeple), NeurIPS Creative AI Track, LA A&D Museum, and more, earning recognition such as the MUSE Design Gold Award (AI Track). Through his practice, Li redefines the boundaries of digital art and human-machine collaboration.
Q: Can you share a bit about your background? What first drew you to media art, CGI, and AI-driven creative projects?
Hi, I am Cory, an AI interface media artist based in NYC.
I was first introduced to media art when I came across the CGI work of Yetong Xin (Reraner) online. Captivated by her visual style and animations, I was inspired to start learning Cinema 4D. This initial spark led me to immerse myself in creating visuals and constructing virtual worlds. Last year, I had the opportunity to collaborate with Yetong Xin on a project called Poespin, which was a fulfilling experience and a milestone in my creative journey.
Another pivotal moment in my path toward AI-driven creative projects came in early 2020. Through Weibo, a Chinese social media platform, I discovered an AI chatbot project called Wander001. This led me to meet Cheesetalk Yuqian Sun, an artist and mentor who has had a profound influence on me. She introduced me to the world of AI art and speculative explorations of language and artificial life. At that time, GPT-3 had not yet gained widespread recognition, but Yuqian’s work was already pushing the boundaries of AI-driven creativity. Her forward-thinking approach inspired me to explore how AI could become a tool to bring my own artistic creations to life.
I spent two years during my undergraduate studies working as Yuqian's artist assistant, which gave me invaluable insights into the intersection of technology and art. That experience cemented my desire to explore the potential of AI as a means to bring my imagined worlds to life.
This impulse—to make my creations feel alive—has driven my artistic practice ever since. It led me to pursue graduate studies at NYU’s Interactive Telecommunications Program (ITP), where I gained the technical skills and creative freedom to bridge my visual storytelling background with AI-driven interactive systems. Now, I use this interdisciplinary approach to turn my creative visions into reality.
Q: How did your experience in New York influence your projects and practice?
Living in New York has profoundly influenced my creative practice in many ways. My project IRL => URL is, in some sense, a tribute to this chaotic city. Two key aspects stand out to me.
First, the people of New York have had a significant impact on me. New Yorkers are incredibly confident and free-spirited—each person seems to have their own anchor in life and a unique pursuit. This environment pushed me to reflect deeply on what I truly want. New York is a place that respects both madness and the necessary dangers that come with freedom. It has clarified my stance on certain issues and my understanding of what freedom means.
Second, on a more direct level, New York is undeniably fashionable. It has an incredible underground culture and a wave of cutting-edge artists. The visual aesthetics of this city have provided me with constant inspiration. I believe these two aspects are interconnected—New York's spirit is not exactly warm, but it is undoubtedly sharp and striking.
Poespin—Wherever your body reaches, there is poetry (2024), Cory Yihua Li, Wendy Li, Jiayi Li, Archy Hongyue Cheng, Reraner Yetong Xin, Armon Naeini, Human-AI Co-created Writing System/ CGI. Image Courtesy of the Artists.
Q: Your work, PoeSpin, merges AI-generated poetry with the physical movements of pole dancing, drawing inspiration from both Chinese poetic traditions and W.B. Yeats. What led you to this intersection of technology, movement, and language?
The intersection of technology, movement, and language in my work Poespin emerges from my broader exploration of media art. Through my studies, I have come to believe that almost everything around us can be translated or collapsed into data—although this translation inevitably results in some information loss. I see AI as a cocreator that attempts to organize and synthesize data in a way that mimics intelligence.
Poespin is my attempt to create a fundamental translation between different forms of expression—bodily movement and language. The project seeks to transform the physical gestures of a pole dancer into readable poetry, aiming to liberate the dancer’s expression while challenging societal biases against pole dance as an art form.
This work operates by unifying various data types—2D, 3D, and 300-dimensional vectors—into a common dimensional space to explore the connections between them. Throughout the process, the roles of AI, the human dancer, and my own decisions as a developer are all indispensable. I view this distributed authorship as an emerging and inevitable trend in media art today. Such collaborative systems raise crucial questions about culture, authorship, and technological nativism.
Ultimately, I hope my work resists the reductionist narratives pushed by capital—those that frame AI solely as a tool for cost-cutting, efficiency, or human replacement. These narratives often exhaust and discourage both creators and the public, and it saddens me to see some people internalize these beliefs, using social Darwinist rhetoric to critique artists. Through Poespin, I hope to show technologists and artists alike that there is an overlapping space where everything can coexist—where technology can add value to human expression rather than replace it.
Q: … and how do you see this work reshaping our understanding of embodied performance and literary creation？
I believe Poespin can be viewed as a continuation or higher-dimensional extension of Oulipo experiments and spatial poetry explorations. Perhaps due to my Chinese cultural background, my understanding of poetry and movement is inherently connected—our classical texts often merge the two. As an artist, I often feel that art is a more advanced and potent form of communication. Immersive performances are not confined to physical spaces; they can digitally explore the connections between various forms of expression. My goal is to create systems where embodied movement, language, and technology converge—revealing new ways to understand performance and literary creation beyond traditional boundaries.
Q: Your projects often integrate AI as a co-creator rather than just a tool. How do you define AI’s role in creative expression, and do you see it as an autonomous collaborator or an extension of human thought?
Given the current state of technology, I lean towards the former view because there are currently no AI models that can absolutely learn human data in real time (this is an architectural issue, and I'm not sure if better AI algorithms will emerge in the future). So perhaps calling it an "extension" isn't quite accurate (though I might change my mind this year, haha). I treat it as a co-creator because, as a non-human intelligence, its unpredictable interactions with users and audiences interest me. Due to how these AI models learn, I don't view prompt engineering results as my creation, but I believe the series of emergent behaviors and experiences triggered by the environment I design are at the core of creativity. However, it's hard to say it won't become an extension of human thought in the coming years. Seeing research from Big tech, and others makes me believe that singularity might be reached, but at that point, the challenges for all artists and creators will be severe. Rather than being passively swept along by technology, artists should actively seek ways to collaborate with it.
Q: By modeling IRL⇒URL after the NYC subway experience, you turn a familiar public space into a metaphor for online interactions and algorithmic guidance. How do you see AI-mediated environments transforming our future social spaces, both online and in physical urban landscapes? Do you see a risk of increased surveillance or control, and how does your work resist or subvert these possibilities?
This is truly a great question! Ultimately, I see my work as opening up these black-box systems for examination. Rather than rejecting AI outright, it's about making its influence visible and playful, so people can engage with it critically. The goal is to empower individuals to navigate these hybrid spaces with greater agency, turning algorithmic infrastructures from invisible constraints into legible playgrounds. While many AI models implement search functions, we must consider the risks of unclear authority and unlimited explorations over user intentions and privacy. Through IRL=>URL interactions, I use special NPCs like drug dealers and thieves to highlight these potential risks within a retro, glitch-aesthetic internet environment.
Q: How do you navigate the ethical and artistic challenges of working with interactive and immersive experiences?
I approach the ethical and artistic challenges of interactive and immersive experiences with a strong sense of curiosity—understanding that I am an explorer in this space, not someone with definitive answers. For interactive and immersive experiences, the most essential and basic thing we should get is the visitor’s consent for using generative models.
Last year, I had the privilege of attending a talk by Ted Chiang, one of my most admired sci-fi writers. He spoke about why he believes AI cannot create art, a perspective he had previously shared in The New York Times. His argument centered on the idea that art requires a deep, iterative process of decision-making, and that this labor—the effort, the struggle, the refinement—is an essential part of what makes something art. However, after discussing some of the workflows used by AI artists today—such as curating datasets, fine-tuning models, and integrating them into creative processes—he acknowledged that this kind of practice could indeed be considered art. But this discussion also highlighted just how ambiguous and difficult it is to define AI-generated art. The term “AI art” itself is misleading—it lumps together vastly different fields, from computer vision (CV) to natural language processing (NLP) to generative models, without clarifying how they are integrated, what the artist’s role is, or what labor was involved in their creation. These nuances are often obscured in mainstream discourse.
That said, I firmly reject the narrative of AI as merely a tool for cost-cutting and efficiency gains, as well as the fear-mongering around it. These are often tactics used by the market to capture attention and manufacture anxiety. As creators, we should redirect the conversation towards where attention should be placed—exploring new ways to co-create with AI rather than being caught up in artificially induced fears or oversimplified narratives.
At the same time, I think decentralized, lightweight generative models are a direction worth looking forward to. Giving creators the ability to customize and have partial ownership over their AI agents presents an exciting vision for the future—one where AI becomes an extension of individual artistic intent rather than a monolithic, corporate-controlled tool.
Q: Your works engage with concepts from Eastern and Western artistic and philosophical traditions—How do you approach cultural hybridity in your work, and how does this synthesis inform your artistic identity?
There is definitely a similar essence in Western and Eastern art practices. The deeper I delve into the field of digital art, the more I sense this connection. Our professor Daniel Shiffman has a book called Nature of Code, which uses programming and mathematical methods to simulate and deconstruct various real-world phenomena, such as force interactions, flocking algorithms, and cellular automata. I find it incredibly beautiful—firstly, because it reveals that nature, at least in its representation, follows discernible patterns, and secondly, because many algorithms remind me of Chinese philosophy.
For instance, in n-body choreography, two-body motion is relatively stable, while three-body motion becomes chaotic and unpredictable. This reminds me of Chinese Taiji yin-yang (and its stability) and Laozi's principle: "Tao creates one, one creates two, two create three, three create everything." I find this beautiful. The East and West have different ways of perceiving the world—China tends toward empiricism, while the West leans more toward logical reasoning—yet they arrive at similar principles and ideas. People's thoughts unconsciously connect and "shake hands" along the veins of knowledge and the world.
My art is a continuous process of calling back to these experiences, exploring different scales and fluidity. I am deeply rooted in traditional Chinese culture and influenced by the cultural aesthetics of both China and Japan. I love calligraphy, Chinese poetry, and painting, and I have always been fascinated by the shared appreciation for nature.
Q: Lastly, any future projects or plans that you’re excited about?
I’m pretty excited about my upcoming thesis—AI Vtubing Creation Platform—and the experimental side I’m building. I’ve always worked independently without much collaboration with artists around me, but after coming to America, I was fortunate to meet creators who truly opened my eyes. They have inspired and helped me in various ways, and I’ve been thinking about what interesting and beneficial things I could do for them.
I’m considering uniting my artist friends and expanding our imagination through Vtubing. I’ll be unveiling some truly unhinged projects related to AI Vtubing—stay tuned.
This open process invites people to witness and engage with the creation journey in real-time. Follow along here: https://x.com/Echuu_AIVTUBING
About Cory Yihua Li
Cory Yihua Li is a New York-based media artist, Vtubing enthusiast, and CGI specialist whose work explores the intersections of AI, digital aesthetics, and interactive storytelling. His innovative contributions have been showcased in prominent exhibitions and events, including the MUSE Design Gold Award (AI Track), NYC ARTECHOUSE, DEJI Art Museum (alongside Beeple), NeurIPS Creative AI Track, Chinese CHI Art Gallery, LA A&D Museum (AI Track), ACM Creativity & Cognition, Tongji Design Week, and the Fukuoka Digital Art Award. Through his practice, Li continues to push the boundaries of AI-driven creativity, reimagining human-machine collaboration in contemporary digital culture.
Website: https://coryleeart.com/
Comments