The Application and Development of Intelligent Art in Contemporary Life

TABLE OF CONTENTS

• ABSTRACT i
• TABLE OF CONTENTS 2
• 1 INTRODUCTION 5
• 1.1 Background 5
• 1.2 Problem Statement 5
• 1.3 Research aims and Objectives 6
• 1.4 Research Questions 7
• 1.5 Significance of the Study 7
• 1.6 Dissertation Structure 8
• 2 CHAPTER 2: LITERATURE REVIEW 9
• 3 CHAPTER 3: METHODS 19
• 4 CHAPTER 4: CASE STUDIES OF INTELLIGENT ART 24
• 5 CHAPTER 5: RESEARCH FINDINGS AND ANALYSIS 28
• 6 CHAPTER 6: DISCUSSION AND ANALYSIS 35
• 7 CHAPTER 7: SUMMARY AND RECOMMENDATIONS 39

1 INTRODUCTION

1.1 Background

The study explains how artificial intelligence (AI) reconfigures creative production by obscuring the distinction between works produced by humans and machines. AI enriches artists' capacities to generate works with minimum intervention through adopting strategies like Generative Adversarial Networks (GANs) and neural networks (Aru, 2025). The technologies question established notions on creativity, authority, and agency. Artists and technologists collaborate with AI to break creative frontiers and test the boundaries of imagination (Galanter, 2016). AI instigates ethical and philosophical concerns around authors' intentionality and intellectual property. Researchers are still in disagreement as to whether works produced by AI signify the progression of creative art or only duplicate prevailing data without ingenuity (Moruzzi, 2025). China is the global pioneer in AI-driven creative innovation. The state encourages the adoption of AI through adopting favorable policy for technology development in the creative industries. China's digital markets welcome works generated by AI in mega-sized events like the Shanghai West Bund Art Fair and the CAFA's AI + Art Exhibition. The study explains how AI is impacting contemporary creative practice, affecting public attitudes, and acting within ethical and legal parameters. An insight into how AI works within the context of Chinese art creates understanding about the broader ramifications of AI-powered creativity within the digital realm.

1.2 Problem Statement

This study examines how artificial intelligence disrupts traditional artistic production. Researchers debate if AI is an artistic tool or a creative autonomous entity. Some scholars believe that AI inspires creativity by offering more artistic options (Goriunova, 2025). Some believe that AI limits genuine innovation by basing it on available data instead of original thoughts. AI-generated work generates legal and ethical issues concerning ownership and intellectual property (Mazzi & Fasciana, 2024). This technology disrupts traditional ownership and copyright protection paradigms. Experts struggle to establish how AI-generated work relates to modern artistic and legal standards (Ameen et al., 2024). This study examines how AI-generated work is received in contemporary Chinese artistic spaces. It examines how AI impacts traditional artistic production and public opinion. The study examines unresolved questions about creativity, authorship, and legitimacy in AI-generated work. Focusing on China, the study provides insights into the intersection between AI, cultural innovation, and ethical concerns during the digital age.

1.3 Research aims and Objectives

This study investigates the impact of artificial intelligence on contemporary artistic practices. It examines AI's role in creative production, audience reception, and ethical considerations.

The specific objectives of this research are:

1. To analyze how AI-generated art redefines traditional artistic practices by influencing artistic techniques, aesthetics, and creative processes.
2. To evaluate the perception and reception of AI-generated artworks among different demographic groups, including professional artists, art critics, gallery curators, and general audiences, with variations across age groups and professional backgrounds.
3. To assess the ethical and intellectual property implications of AI-generated art, particularly regarding authorship, originality, and ownership rights.
4. To explore how AI drives innovation and experimentation in various artistic disciplines, including visual arts, performing arts, and interactive media.
5. To examine the role of China as a leading hub for AI-driven artistic innovation by considering government policies, the digital art market, and institutional integration of AI-generated art.

1.4 Research Questions

1. How does AI-generated art redefine traditional artistic creation, particularly in terms of artistic techniques, aesthetics, and creative authorship?
2. How do different stakeholder groups—including professional artists, art critics, gallery curators, and general audiences—perceive AI-generated art, and how do these perceptions vary across demographic factors such as age, profession, and experience with digital art?
3. What ethical and legal concerns arise regarding authorship, originality, and intellectual property rights in AI-generated artistic works?
4. How does AI contribute to innovation and experimentation in various artistic fields, including visual arts, performing arts, and interactive media?
5. Why is China emerging as a key player in AI-driven artistic innovation, and how do government policies, the digital art market, and institutional frameworks influence the adoption and reception of AI-generated art?

1.5 Significance of the Study

This study takes into account the evolving dynamic between artificial intelligence and creative production. This study takes into account how AI-generated content challenges the conventional understanding of authorship, originality, and creative agency. This work provides policymakers, curators, and creatives with insight into the challenges and opportunities posed by AI in creative fields. This work offers ethical implications and regulation suggestions to bring AI into artistic production. This work emphasizes inter-disciplinary collaboration between creatives, technologists, ethicists, and policymakers. This work is striving to encourage cross-disciplinary conversation to make sure that AI is responsibly incorporated into artistic production while ensuring cultural authenticity and artistic integrity. This work is also educating educational and cultural institutions to adopt best practices to bring AI into creative pedagogy. This work also makes sure that AI is used as an addition to, and not a substitute for, human artistic production.

1.6 Dissertation Structure

The seven chapters in this dissertation include Chapter 1, which introduces the problem, purpose, and significance of the study. Chapter 2 is an extensive literature review, summarizing theoretical understanding into AI and art, the background, and key arguments on authorship and ethics. Chapter 3 is a description of qualitative research methods, including case studies, interviews, and field observation. Chapter 4 is a reporting of findings from selected case studies on AI-generated artworks in China, summarizing their impacts on artistic practices. Chapter 5 combines research findings from interviews and field observation to examine public reception and ethical concerns. Chapter 6 interprets these findings against literature, bridging research gaps and theoretical understanding. Chapter 7 concludes the study by summarizing key findings, offering policy recommendations, and suggesting avenues for future research.

2 CHAPTER 2: LITERATURE REVIEW

2.1 Introduction

The literature review is based on historical development, developments in technology, and contemporary application of artificial intelligence in art. The section critically examines prominent academic works on authorship, creativity, and ethics and assesses machine learning, neural networks, and generative adversarial networks' contributions to digital and interactive art.

2.2 Theoretical Foundations of Intelligent Art

The deep learning algorithms and other complex algorithms are capable, nowadays, of producing paintings, music, texts, and interactive performances that challenge long-held ideas about authorship; thus, the artificial intelligence is perceived as a revolutory force that will change the very essence of human expression. According to Vinchon et al. (2023), what AI is, in reality, is not an isolated creative force, but rather a cooperative instrument that enables one to express his or her human imaginative capacities.. Ameen et al. (2022) point out that machine learning algorithms such as Generative Adversarial Networks (GANs) are behind making creativity in AI a reality through large databases. This is countered by Aru (2024) who is of the opinion that AI is not self-motivated and is not purposeful in actions and hence cannot create in a human sense. Grilli and Pedota (2024) further argue that works by AI are pattern-based and not based on actual cognitive originality and hence are a question as to whether such art by AI is nothing more than a statistically recombination of existing material and not a product of creative thinking. These arguments raise an open question: while indeed art possibilities are being extended by AI, questions are raised as to whether it is a complete independent creative force or is only a powerful computation tool.

Computational creativity is a conceptual framework that is used in assessing art created by AIs and whether it is capable of creating new ideas independent of itself or only mimicking patterns that have been learned. Ameen et al. (2024) explain that creativity in AIs is measured by whether it is capable of creating novel and valuable and surprising results, subjective measurements to approximate. Moruzzi (2025) criticizes Margaret Boden's creativity theory as combinational, exploratory, and transformational and theorizes that AIs are perhaps excellent at combinational creativity as a question of combining styles of art that already exist. However, Tigre Moura et al. (2023) discovered that aesthetically beautiful art created by AIs is still favored by audiences because it is perceived as having more depth and authenticity by human standards. In similar context, Al-Rousan et al. (2025) theorize that AIs are capable of creating technologically superior works of art but that AIs are not conscious and are not contextualized in culture and as such their works are inherently not comparable to that created by human beings. These arguments theorize that AIs' ability as a creative agency is not yet solved since their works are still very much dependent on human intervention and training.

2.3 Historical Development of AI in Art

The roots of AI art date back to the earliest forms of algorithmic art that existed before the advent of modern-day deep learning. Computer scientists and artists in the mid-20th century experimented with rule-based computational systems to create visual art. Kafer and Quick (2021) note that the earliest computer art methods mainly involved algorithms working with geometric shapes and abstract patterns, paving the way for the current AI art. Harold Cohen was one of the prominent figures of the period who came up with AARON—a revolutionary computer system that could create unique drawings on its own (Sundararajan, 2014). AARON differed from current-day AI models that use deep learning in that it operated with predefined heuristics to create unique artwork. Although the initial systems showcased machine creativity, they did not have the autonomous learning aspect that exists in modern-day models (Bertram, 2017). The transition from algorithmic art to AI art is a development that moved away from the use of rules in programming to adaptive machine learning models, changing the way AI is applied to creativity in art.

The advancement in machine learning and neural networks has revolutionized art produced by artificial intelligence by enabling models to learn from vast data and generate more complex art. Moruzzi (2025) explains that with the advent of deep learning algorithms such as Generative Adversarial Networks (GANs), creative use of artificial intelligence has been revolutionized by making it possible for machines to create new art styles on their own. A breakthrough came with the advent of Google DeepDream that used convolutional neural networks to produce dreamlike surreal images by enhancing patterns in existing pictures (Goriunova, 2025). The process differed quite a lot from algorithmic ones in the past since the results of DeepDream were not deterministic and were built dynamically through iterative neural computation. In a similar context, with DALL·E and Midjourney, artificial intelligence's capability to produce very detailed and rich conceptual art based on text prompts further blurred human and machine creativity. Even though these developments demonstrate more refinement in art with artificial intelligence, they introduce new challenges in originality, authorship in art, and ethical implications in creativity with AI.

Throughout history, AI art has made a series of significant milestones and with that has marked progress in computational creativity. One of its earliest breakthroughs came in Harold Cohen's AARON when despite its use of predetermined rules, it laid down the groundwork of machine-generated art (Sundararajan, 2014). Contrary to this, today's AI models utilize large-scale data and adaptive learning to facilitate very detailed and context-aware art. The release of Google DeepDream in 2015 created a turning point by demonstrating that neural networks could be used to venture into art in a new and creative way (Goriunova, 2025). Recently, DALL·E and Midjourney have pushed limits of AI-generated creativity by allowing users to generate photorealistic and abstract images based on text descriptions and effectively democratized artistry with AI assistance (Mazzi & Fasciana, 2024). Legal and ethical implications of these are yet to be agreed on and pertain to intellectual property and human intervention in works created by AI (Bertram, 2017). With advancements in AI moving apace, only one question remains to be answered: whether it will be a cooperative tool for human artists or a force that subverts customary notions of authorship in art.

2.4 The Economic Expansion of AI in the Art Market

The increasing adoption of AI in artistic creation has not only transformed creative processes but has also led to significant economic growth in the art market. As AI-generated art becomes more integrated into digital platforms, galleries, and institutional collections, its market value has risen substantially. According to Shalwa (2024), the global AI in art market is projected to expand from $3.2 billion in 2023 to approximately $40.4 billion by 2033, reflecting a Compound Annual Growth Rate (CAGR) of 28.9%.

2.5 Key scholarly debates: AI's role in authorship, creativity, and ethics

Copyright law traditionally respected authorship by natural persons, but the introduction of AI-generated content puts the legal basis under siege. The courts in most jurisdictions have decided such matters differently. The Thaler v. Perlmutter (2023) case is a classic example of the inflexibility of U.S. copyright law since the court refused to grant copyright protection to a work produced by an AI program, affirming the stance that authorship must be attributed to a natural person (Thaler v. Perlmutter, 2023). Under Japanese law, copyright law was modified to enable the use of copyrighted content for the purpose of training AI but does not protect the output of AI (National Law Review, 2019). The above legal positions demonstrate reluctance in adapting authorship with the advancement of technology. Although such judgments consolidate legacy legal precedents, the approach is lacking in addressing the larger role now undertaken by AI in creative sectors. Withholding copyright protection from content produced by AI raises pragmatic issues, most notably with respect to ownership, economic incentives, and creative accountability. If content produced by AI is confined to the public domain, then creativity by AI may be appropriated by corporations without the compensation payment by the original content providers or developers of the AI program. The above inflexibility may dampen creativity within the use of AI-based creative sectors, and a wiser approach is necessary.

Courts across jurisdictions have rendered conflicting verdicts on AI-generated content, and therefore there is ambiguity regarding copyright law. The case Sarah Andersen et al. v. Stability AI Ltd. in the United States is an indicator of the emerging tension between copyright law and AI-generated content, as Stability AI is charged with unauthorized use of copyrighted work (Andersen, McKernan, & Ortiz, 2023). In the UK case Getty Images v. Stability AI, Getty alleged that Stability AI's data scraping was an infringement of their copyright (Getty Images v. Stability AI, 2023). In China, courts have been inconsistent: Feilin v. Baidu declined to accord copyright protection to AI-generated work, but Tencent Shenzhen v. Shanghai Yingxin recognized AI-assisted content as protectable under China's copyright law (Lee, 2021). The legal fragmentation creates ambiguity for content providers and AI developers, and there is a gap in the current copyright systems. The inconsistency means that courts struggle to draw the line between AI assistance and authentic AI authorship. Rather than asking if AI can be an author, a more practical legal system should decide the extent to which humans contributed to AI-generated work and attribute the copyright accordingly. Consistency in copyright policies across regions would provide more clarity to AI-related content production and commercialization.

Ethical debates surrounding AI-generated art extend beyond copyright disputes, raising concerns about originality, fair compensation, and the integrity of human creativity. AI-generated content relies on vast datasets, often incorporating existing copyrighted works without explicit authorization. Critics argue that this process dilutes originality and homogenizes artistic expression (Gao, Raess, & Zeng, 2023). Additionally, corporate AI applications, such as Stability AI's models, prioritize mass production of AI-generated content at the expense of human artists (Getty Images v. Stability AI, 2023). This trend has sparked fears that AI-generated art will replace human creators, diminishing the value of human artistry in digital media. While AI undeniably enhances creative possibilities, its growing dominance threatens the diversity and originality of artistic expression. If legal frameworks fail to regulate AI's use of copyrighted material, human artists may face economic marginalization. Ethical AI governance should ensure transparency in AI training datasets and establish fair compensation models for original creators whose work contributes to AI-generated content. Addressing these ethical concerns is critical to maintaining a balanced creative ecosystem that benefits both AI developers and human artists.

The legal and ethical debate surrounding AI-generated content is one of ownership—whether to give copyright protection to these works, or to put them into the public domain. Some policymakers and academics suggest keeping AI-generated content in the public domain, as AI is not attributed to have creative agency and should not be given exclusive rights (Gao, Raess, & Zeng, 2023). Some suggest other frameworks of ownership, such as granting copyright to AI developers or to individuals contributing creative input through AI-generated prompts (Lee, 2021). However, the lack of clear legal precedent makes it problematic to determine how to categorize AI-generated content under current copyright law. Placing AI-generated content in the public domain may foster knowledge-sharing, but may also motivate large corporations to profit from AI-generated content without credit or compensation. A fairer solution would be to implement a hybrid system of copyright, treating AI-assisted works as co-creations where human input is a deciding factor in authorship. Implementing an AI-specific intellectual property system would provide legal certainty and reward technological innovation as much as it would reward human creative input.

2.6 AI in Art Education

Artificial intelligence is revolutionizing the study of art, impacting pedagogy and the learning of skills. Applications based on AI, including DALL-E, DeepDream, and Midjourney, introduce new ways to generate visual work, and machine learning enables interactive learning in music and creative writing (Mazzi & Fasciana, 2024). AI-based creativity in learning defies conventional wisdom regarding the learning of techniques, as students can generate complex pieces without comprehension of the techniques used. However, critics argue that overreliance on AI can compromise basic creative skills, including manual drawing and thinking concepts (Al-Rousan et al., 2025). The introduction of AI to learning is both an opportunity to widen artistic horizons and a problem in maintaining basic creative capacities. While AI opens learning to the arts, there are ethical concerns regarding originality and plagiarism. The ability of AI to rearrange large sets of data into work creates intellectual property concerns regarding the ownership and originality of work by students utilizing AI (Thongmeensuk, 2024). With more AI tools found in classrooms, teachers have to balance innovation with ethical concerns, ensuring that AI enhances and not hampers creativity.

2.7 Ethical and Legal Implications: Japan's 2023 AI Copyright Guidelines

Japan's 2023 copyright law regarding AI reflects an evolutionary legal system to address the complexities of AI-generated content (National Law Review, 2019). The General Understanding by the Japan Copyright Office regarding AI and Copyright in Japan elucidates the application of Article 30-4 under the Japanese Copyright Act, under which training AI on third-party content is permitted under certain conditions. Specifically, the law permits AI models to use copyrighted content as long as the use is not to "enjoy the thoughts or sentiments" contained in the works (Japan Copyright Office, 2024). This distinction creates significant legal controversies: if AI-generated content is transformative or merely derivative, and to what extent training data infringes on original copyright holders' rights (Gao, Raess, & Zeng, 2023).

One major prohibition under Article 30-4 is that AI training must not use explicitly constructed datasets to generate content similar to content in copyrighted materials (Japan Copyright Office, 2024). This aligns with recent global litigation, including Getty Images v. Stability AI (2023) in the UK and US, where plaintiffs argue that AI models violate protected images by scraping and repurposing them without licensing (Getty Images v. Stability AI, 2023). Another provision under Japan's law is that AI training must not unfairly prejudice copyright holders, setting precedent for AI companies to get equitable licensing agreements rather than taking creative work without compensation (National Law Review, 2019). Japan's approach is a middle ground between lax AI use and strict enforcement of copyright, stressing global harmonization of AI copyright law. Japan's AI copyright law offers strong legal protection but is still favorable to AI innovation. However, critics argue that the "non-enjoyment" provision is too ambiguous, as AI-generated content will reflect artistic intention by humans (Gao, Raess, & Zeng, 2023). The law attempts to prevent copying but is not strong enough to respond to the ethical dilemma posed by AI's effect on creative industries. As AI-generated artwork gains traction, Japan's forward-looking but conservative approach is an exemplar to other nations seeking to balance technological innovation with intellectual property rights.

2.8 Technological advancements: Machine learning, neural networks, GANs

Machine learning and neural networks have successfully revolutionized artificial intelligence with support for breakthroughs in automation, handling of data and predictive modeling. Rule-based and static systems used by classical AI were inadequate at handling unexpected situations. The introduction of neural networks in the form of deep learning models made it possible for AI to mimic human thinking by recognizing intricate patterns in large volumes of data (Tiwari, 2024). Convolutional neural networks revolutionized AI's image processing capability and recurrent neural networks improved handling of sequences that has made breakthroughs in language translation, medical imaging, and real-time analytics (Wang et al., 2023). These developments come at a cost—deep learning models are compute-hungry and consume vast amounts of computation and are ravenous in their demand for data, posing ethical concerns over biased data samples, privacy invasions, and environmental harm through excessive power usage (Masood et al., 2024). Large-scale deployment of training data brings in a kind of imbalance where only tech giants with humongous resources are in a position to harness the power of AI to its full potential and further concentrate power in the industry.

One of the most profound breakthroughs in AI research has been the development of generative adversarial networks (GANs), revolutionizing synthetic data synthesis, content generation, and security use cases. GANs learn through an adversarial cycle between the generator and discriminator and generate very realistic images, sound and even deepfake video (Singh et al., 2024). Such use cases have revolutionized industries such as pharmaceuticals where GANs are used to generate synthetic medical data to conduct research on diseases and drug repositioning (Arjmandi-Tash et al., 2024; Meenakshi et al., 2020). Cybersecurity is another domain where GANs have found use with new ways of cryptographically generating keys and data security (Diwan et al., 2025). However, one cannot ignore the dark implications of GANs. The propagation of AI-generated media has been a source of concern regarding misinformation and propagation of false identities and public distrust in digital information (Verma et al., 2025). Adversarial actors have used deepfake tech to manipulate political facts and deceive customers and undermine credibility in journalism and media reporting (Yingngam, 2024). Lack of regulation allows AI-generated material to spread at a frenzied pace and leave attempts at developing detection tools in its wake, raising urgent questions regarding accountability and digital ethics.

Aside from GANs, advances in semi-supervised and unsupervised learning are increasingly opening up opportunities in both creativity and automation using AI. While deep learning used to need huge annotated data, more recent architectures have self-learning capabilities and therefore are less dependent on human-labeled data sets (Masood et al., 2024). Such developments are particularly useful in areas such as healthcare, where AI systems now read complex biological signals, predict patterns of disease development and guide personalized medicine (Verma et al., 2025). With such and other applications, however, come risks of substitution by AI automation of human expertise. Increased reliance on machine-generated insights necessarily threatens explainability, since black-box systems often make decisions without clear rationale (Diwan et al., 2025). With increasingly autonomous AI systems, moreover, legal and ethical bases of accountability are questionable—if an AI-assisted medical diagnosis is wrong, who is responsible: developer, healthcare provider, or the AI itself? Such challenges demand interdisciplinary collaboration among technologists, ethicists, and policymakers so that AI development occurs not only innovatively but socially responsibly.

2.9 Research Gaps

Notwithstanding sudden breakthroughs in AI-generated art, research lacunas are significant. Ethical and legal uncertainties, in particular in authorship and copyright issues, are prevalent as jurisdictional powers differ in dealing with AI-generated art with uncertainty regarding property and creative rights. The creative agency of artificial intelligence is limited as well—while AI is capable of creating stunning visual and aural works, it has no real human intuition, cultural awareness, and depth of emotion-based intentionality and questions its ability to produce meaningful and contextually relevant art. Inadequate cross-disciplinary interaction between researchers in AI and artists and ethicists is also slowing down formulating frameworks that have AI as a co-creative partner rather than a content provider. Lastly, technological constraints such as biased data in AI training and energy use in deep learning frameworks hamper AI's ability in creating authentically original works and present sustainability challenges for future creativity breakthroughs through AI.

4 CHAPTER 4: CASE STUDIES OF INTELLIGENT ART

4.1 Introduction to Case Studies

Use of artificial intelligence in contemporary art has grown exponentially, particularly in China where AI-created creativity is revolutionizing art production. The chapter offers case studies of the use of AI in various types of art and offers an insight into technology and creativity and how they intersect with culture.

4.2 Case Study 1: AI in Visual Art

The advent of AI-generated visual art has revolutionized creative expression and challenged human and machine-generated aesthetics. The most well-known works of art include "Portrait of Edmond de Belamy," created by Parisian group Obvious through a Generative Adversarial Network (GAN) that had been trained on a 15,000-piece collection of classical portraits (Aru, 2025). The sale at Christie's for $432,500 sparked heated debate regarding authorship, originality, and where art and AI fall. The AI program designed to identify patterns in historical art created a unique and innovative piece that as aesthetically captivating as it may be lacked human intentionality in creativity (Galanter, 2016). Critics argue that AI-generated art only reproduces pattern and not creative intention and pose questions on its ethical implications on traditional art practice (Mazzi & Fasciana, 2024). Other artists and researchers view AI as an augmentative and not a substitute tool that comes with novel possibilities of human-intelligent system collaboration (Tigre Moura et al., 2023). Popular opinion regarding AI-generated visual art has been mixed as well with others embracing it as a thrilling innovation and others doubting its legitimacy and quality as art. In spite of such contention, AI-generated art has found traction in commercial, academic, and exhibitary spheres of interaction and has brought with it its influence on contemporary art culture.

Figure 2: AI-Human Collaboration in Visual Art, which visualizes the process of integrating AI in art creation. It clarifies the AI-human workflow, showing how artists contribute at various stages.

4.3 Case Study 2: AI in Performing Arts

The integration of AI in performing arts has transformed creative processes, particularly music composition and dance choreography, with AI systems now assisting in creating new forms of art. A good example is "AI Duet," an interactive performance developed by Google's Magenta project that enables musicians to collaborate with an AI-powered system that has been trained on neural networks and recurrent learning algorithms (Aru, 2025). Another example is that of Wayne McGregor's collaboration with AI using machine learning algorithms to process thousands of dance movements and generate new choreographic sequences that showcase AI as a potential collaborator and not a replacement for human creativity (Goriunova, 2025). The extent of human-AI collaboration varies—some performances rely heavily on AI-generated content, whereas others use AI as an augmentative component to facilitate human performance. Audience reaction has been mixed with some embracing the potential of AI to extend creative horizons and others criticizing its lack of authenticity and emotional depth in live performances (Tigre Moura, 2023). The industry experts and critics are also cautious with their acceptance as they remark on AI's ability to generate technically sound compositions and point towards its lack of human intuition and spontaneity that form essential parts of live performances (Grilli & Pedota, 2024). In spite of these challenges to accepting AI in performing arts, its use continues to rise with new possibilities presented in adaptive performances in real-time, interactive audience engagement, and interdisciplinary collaboration. However, a lack of originality and in-depth emotional intelligence remains a limiting aspect that has raised questions regarding the ethics and viability of AI in live art.

Three significant Chinese AI art exhibitions—Shanghai West Bund Art Fair, CAFA's AI + Art Exhibition in Beijing, and Shenzhen Smart Art Expo—offered perspectives on how the public responded to performances by AI. They were mixed in their reactions, varying from excitement and interest in the utilization of AI-generated performances to distrust in the artistic value of AI. Whereas some attendees welcomed the utilization of AI as a creative entity, others doubted its capacity to express true artistic feeling.

5 CHAPTER 5: RESEARCH FINDINGS AND ANALYSIS

5.1 Introduction

This chapter provides an in-depth summary of findings from fieldwork, interviews, and case studies of artwork produced by artificial intelligence. In this chapter, the author explores the intricacies of creativity, authorship, and audience experience in the context of visual and performance arts as they relate to AI. The findings have then been framed in three separate sections that highlight these essential themes: innovation, ethics, and public reception.

5.2 Case Study Findings

5.2.1 AI in Visual Art

The first case study explored the way artificial intelligence (AI), as embodied by a sophisticated generative adversarial network (GAN), generated visual art. Through the study of a vast number of historical examples of paintings run through machine learning algorithms, the resulting art revealed the level to which these algorithms have evolved to mimic traditional methods of art creation and demonstrate a high level of creativity. The second case study also explored the way that AI was trained by way of deep learning models to come up with unique composition schemes. The resultant artwork generated caused a wide range of responses among critics as well as the general public: whereas a portion of it recognized its potential to unlock future creativity, others concerned the issue of its authenticity as well as the position of human artists in the world of art generated by AI. The issues of authorship and intellectual property connected to these new creations challenged the most fundamental beliefs regarding the nature of the origin of art. The case study fully explored the changing dialogue between visual art and artificial intelligence, with the potential for innovation versus the question of the artist's place in an increasingly digitized world.

5.2.2 AI in Performing Arts

The primary case study presented the examination of AI's very interaction with the vision of painting, where a GAN in itself constituted the very formula of the making of the artwork to be discussed. The artwork, whose lifeline extended from an AI-trained multi-century graphical history, exemplified the anticipatory talent machine learning algorithms have to iconically replicate or innovate upon art traditions. This study paid especial close attention to the AI learning process along with the possibilities deep learning models have to create new compositional families. It evoked mixed reactions from art critics and the public; while some celebrated the talent of AI helping broaden new avenues of creativity, others questioned its authenticity and the role artists were to play within AI art. Ethical implications were posed through questions about authorship and intellectual property as the AI-generated artwork was seen as challenging traditional notions of originality. The case study investigated the interplay between AI and visual arts by further illuminating both potential avenues of innovation and other uncertainties regarding artistic agency in the digital environment.

5.2.3 AI in Literary Arts

The third case study involved AI in interactive media with particular attention to an AI-facilitated immersive art exhibit that responded in real-time to audience interaction. The exhibit used machine vision and machine learning algorithms to detect motion, emotion, and gesture and therefore enable visual and sonic responses to be changed by the AI program. The interactive and adaptive approach redefined traditional viewer interaction as spectators became active participants and contributors. While the exhibit received praise as a dynamic and personalized art experience, its application brought with it questions regarding information privacy and ethics in AI use in monitoring human activity. Moreover, research brought into question whether responses by an AI can be termed as "artistic intent" or were merely computation. Together, research illustrated that AI in interactive media unleashes creative potential through personalized and immersive experience yet with increased participation come ongoing questions regarding ethics, authorship, and viewer agency.

5.3 Interview Findings

5.3.1 Human-AI Collaboration in Art

Out of the 70 interviewees, 41% viewed AI as a tool that complemented rather than replaced human creativity, as illustrated in Figure 3. These respondents described AI as a means of expanding artistic possibilities, particularly in its ability to generate unexpected patterns and compositions that artists refined and adapted. However, 10% of respondents—primarily curators and traditional artists—expressed concerns about AI's impact on authorship, arguing that AI tools automated key aspects of creative decision-making, thereby diminishing the artist's role in producing original work. Additionally, 30% of participants noted that AI-generated art tended to reinforce dominant artistic styles, as models trained on large datasets replicated existing trends rather than introducing new creative directions. This raised questions about whether AI expanded artistic horizons or merely mimicked prevailing aesthetic preferences. Furthermore, 19% of respondents remained skeptical about AI's legitimacy in artistic creation, asserting that AI-generated works lacked the emotional depth and intentionality associated with traditional human-made art.

5.3.2 The Future of AI in Artistic Creation

Out of the 70 interviewees, 31.2% supported AI as a creative tool, believing it functioned as an extension of artistic practice rather than an independent creator, as shown in Figure 4. However, 28.2% of respondents expressed hesitation regarding copyright issues, arguing that AI models, trained on thousands of copyrighted artworks, produced derivative rather than original content. The issue of authorship remained unresolved, as 22.4% of participants debated whether ownership should belong to the programmer, the artist who trained the model, or the AI system itself. Additionally, 18.2% of respondents observed bias in AI-generated content, noting that AI models, trained on existing datasets, reflected the systemic biases of the data they were exposed to, often favoring dominant artistic trends.

5.3.3 Market and Institutional Adoption

Out of the 70 interviewees, 37.3% supported AI as a creative expansion, believing it offered new artistic possibilities and challenged conventional artistic boundaries, as shown in Figure 5. However, 26.4% remained skeptical about AI-generated art, arguing that AI's role in art-making compromised traditional values of craftsmanship and originality. Additionally, 16.4% of respondents questioned AI's artistic legitimacy, asserting that art required intentionality and human agency, qualities they believed AI lacked. Meanwhile, 20% of interviewees preferred AI-generated art when it incorporated interactive elements, observing that audience engagement increased when AI art was participatory rather than static.

5.3.4 Style Homogenization in AI-Generated Art

A significant portion of respondents, 30.9%, observed that AI reinforced dominant artistic trends, rather than fostering creative originality, as illustrated in Figure 6. These participants noted that AI models, particularly platforms like Midjourney and DALL·E, repeatedly generated highly stylized visuals that followed popular digital aesthetics, making AI art more derivative than innovative. Furthermore, 27.6% of respondents expressed concerns about originality, arguing that AI's reliance on pre-existing datasets restricted its ability to produce novel and distinctive artistic expressions. Additionally, 23.0% of participants found AI-generated works highly polished but formulaic, suggesting that AI's efficiency in refining artistic elements came at the cost of unpredictability and genuine artistic experimentation. However, 18.4% of respondents believed AI could innovate new styles, particularly when used as a collaborative tool where human artists guided and refined AI-generated compositions.

5.4 Field Observation Findings

5.4.1 Audience Engagement with AI Art

Through fieldwork observations, significant insights into AI integration within artistic spaces were collected. Observing AI-enabled visual art exhibitions, it was noted how audiences interacted with these artworks in stark contrast to traditional exhibits—many viewer types have exhibited long durations interacting with the pieces, often with the assistance of augmented reality (AR) or interactive digital screens. One curious visitor asserted: "Fascinating to think this work was produced by an algorithm, not a human artist. It really questions what we consider authentic creativity." Mixed responses from audiences toward AI-enhanced performances were noted; traditionalists grew more skeptical as the endeavor played out, especially when it was AI that generated music or choreography, while younger viewers expressed elation. AI-generated dance moves seem broad yet futuristic, opening up new creative possibilities, one boasted. Performers evinced contradictory emotions: helped by AI in expressing their creativity, but equally apprehensive as AI was assumed to be a serious threat to the creativity of human performers. Lastly, intelligent art installations built in public spaces manifest how AI-dependent exhibits promoted engagement: many individuals interfaced with installations by way of providing inputs changing the work in front of them. "This is way more immersive than static painting — it's like art speaks to me," another of the participants quoted. Such findings suggest that modes of art same with AI not only redefine artistic production but also heavily influence audiences in suppressing ways regarding their reception and participation.

5.4.2 The Impact of AI on Artistic Techniques

Using field observations, one found art installations created with the use of AI in various exhibit sites that ranged from galleries to public areas. Many of these had interactive features where audience input played a primary role in deciding on the final art piece. In one of the observed exhibits, for example, an art installation created by an AI changed in response to audience movement and changed its visual pattern in real-time. The interactive and experiential nature of such art projected AI as being capable of changing and dynamic art that challenges static forms of conventionally conceived art. The curators further indicated that more individualized art experience is achievable where visitors have customized interactions with art. Other visitors were critical and indicated that art created with the use of AI did not have intentionality and depth of emotion like artwork by humans. This polarization in audience response addresses the controversial issue of AI as a creative force in itself versus a means of enhancing art.

6 CHAPTER 6: DISCUSSION AND ANALYSIS

6.1 Introduction

Here, findings in Chapter 5 are critically compared with what has been found through a review of literature. The aim is to identify key alignments, contrasts, and lacunae between current theory and evidence found in case studies, interviews, and field observations.

6.2 Comparative Analysis of Case Study

The examination of AI-generated visual and performance art in the case studies aligns with existing literature in illustrating AI's capacity to generate creative works that are on par with those created by human artists. Scholars like Galanter (2016) refer to the contribution of generative adversarial networks (GANs) in broadening art potential, and indeed such broadening is found in AI-generated visual art examined in the present study. Likewise, Goriunova (2025) is interested in emphasizing participatory platforms in AI-driven arts, and indeed so with interactive elements in AI-generated performance found from field research. Nevertheless, a main gap lies in examining originality and authorship. Although theory in literature like Mazzi and Fasciana (2024) discusses legal potentials in AI-generated works, the case studies showed that most artists still remain ambivalent regarding rights of authorship. According to one interviewee, "AI enhances my creativity, but I am not sure if I can fully claim my work as mine." Such ambivalence reveals that although theory in AI-generated art has gone far, pragmatic mechanisms to deal with authorship are still in its infancy.

In addition, in accordance with literature, art is democratized by AI in that it empowers non-professionals to harness powerful tools (Aru, 2025). Contrary findings are presented in the case studies in that respondents made mention of a steep learning curve with respect to art created by AI. In one artist's opinion, "Mastering AI tools takes time. It's not as easy as clicking a button." This is evidence that theory and practice have a discrepancy. Another issue of contention in conceptualization is collaboration with AI. Literature highlights AI as a joint creator and not a standalone creator (Tigre Moura et al., 2023), yet in the case studies carried out it happened that a leadership position in creativity took place in which a disagreement occurred regarding replacing and not complementing human creativity.

6.3 Comparative Analysis of Interview Findings

The findings of interviews supported several observations from theory and literature as well as a number of key discrepancies. Existing research, such as that of Mok et al. (2025), argues that AI stimulates creativity since it brings about new tools for creative expression. This was supported by several of those who participated in interviews and acknowledged AI's role in opening up their creative process. As one interviewee noted, "AI is like a helper that assists me in coming up with ideas that I would not have come up with otherwise." Contrary to that of Mok et al., however, several participants in the interviews questioned overuse of productions made by AI at the cost of losing personal artistry.

A second significant area of discrepancy pertains to the impact of AI on arts accessibility. The research and Ameen et al. (2024) reveal that AI makes art more accessible and increases access. Yet, interview findings gave a more mixed picture. Although some respondents assumed that AI democratized access to art, others noted that cost and technical ability were a barrier. An artist provided a clarificatory remark: "AI tools are not always in budget and without adequate training, it is a challenge to be able to make much use of it." This variation suggests that although AI has potential to be more inclusive, practical barriers are in the path of its full realization.

6.4 Comparative Analysis of Field Observations

Field observations provided thoughtful information on AI art making that supported some current research and, in certain instances, provided unexpected inconsistencies. Studies such as Galanter (2016) argue that art making is being revolutionized by AI-generated art because it makes art possibilities available that are outside human capabilities. Observations at field sites on AI-generated art provided evidence for this phenomenon by illustrating depth and creativity that went beyond art's norm. Such an exhibit included an AI that changed its visual patterns in real-time as it responded to audience input and provided evidence of AI's capability to produce dynamic art.

Yet romanticized in the literature is AI's promise to democratize creativity. Goriunova (2025) highlights that open platforms have made it easier for more people to participate in creative art. Though observations at interactive exhibitions of AI art confirmed more participation, barriers were exposed. Most visitors found it challenging to use technical interfaces, illustrating that despite AI's promise, access is an issue. Additionally, although works like Ameen et al. (2024) acknowledge that AI has the potential to minimize creative constraints, observations in the field found that many artists struggle to integrate AI seamlessly into their creative process. Multiple artists at the exhibition commented that AI-generated content had to be heavily edited by humans to be presented as art, contradicting belief that AI fully automatizes art creation.

6.5 Theoretical and Academic Contributions and Implications

The research contributes to existing theoretical discourse in AI-generated art by creating a bridge between conceptual theory and empirical practice. The research is in line with and complements existing theory as in Galanter's (2016) theory of art generation that holds that AI redefines creative practice by bringing on board autonomous systems that are creative in making unique works. While this found support in field observation and case studies, research is found to show a prominent human-AI cooperation dimension that theory downplays. In opposition to theory that assumes that AI works in isolation as a creative force, empirical observation is found to indicate instances of intervention by artists in artifacts generated by AI to redefine and refine them. This is a pointer towards evidence of hybrid nature of AI art and necessitates a new theory that accepts a marriage of human and machine intelligence.

Academically speaking, this research empirically substantiates Ameen et al. (2024)'s assertion that AI enhances product and service creativity. However, while their research highlights the efficacy that AIl brings to creativity, in the case studies it became apparent that AIl doesn't so much eliminate creative constraints as relocate them. Instead of more-traditional skill and technique-based constraints, artists now have to deal with technical and conceptual obstacles, such as model training and ethical considerations. This leads one to conclude that future research would be well advised to look at what new skills are required in AIl-enabled creative practice and what needs to be done by art school in response.

7 CHAPTER 7: SUMMARY AND RECOMMENDATIONS

7.1 Summary

The aim of this research was to explore artificial intelligence as used in contemporary art practice with specific focus on its application in visual arts, performance arts and interactive installations. The research questions were to explore ways in which AI is changing art making, its impact on authorship and creativity and audience and industry reception. These questions were explored through qualitative research approaches like case studies, artist and developer interviews and observation of art exhibitions that utilize AI. Findings indicated that AI has become a copartner in art making with tools like generative adversarial networks and deep learning systems enabling creation of new art content. In visual arts, AI has brought new digital aesthetics with creativity and originality being questioned. In performance arts, AI-generated composition and choreography have enhanced human creativity as opposed to replacing it and resulted in fusion art forms. Additionally, interactive installations with AI-generated content have enhanced audience experience with art and made it more interactive and engaging. However, critical issues emerged with most critical ones being those that border on authorship, intellectual property rights and ethics. The research found that as art made by AI grows in popularity, there is ambiguity regarding its authorship. The analysis went on to illustrate ways in which AI as a facilitator is at loggerheads with its potential as a disruptor in art making. These findings validate those in literature but illustrate lacunas in ethical standards and regulatory policies on art made by AI.

7.2 Recommendations

1. Governments and legal institutions should adopt standardized policies, such as the EU's 2024 AI Art Copyright Directive, to define authorship rights and equitably reward both AI developers and artists.
2. Cultural institutions and funding bodies should launch AI-art grants and residency programs to encourage collaboration between artists and AI engineers, fostering creative augmentation rather than replacement.
3. Research institutions and AI developers should prioritize the creation of energy-efficient AI models, reducing the environmental impact of high-energy-consuming AI systems like GANs.
4. AI developers and art institutions should implement ethical AI training practices by ensuring dataset transparency and promoting diverse artistic representations to prevent cultural bias.

REFERENCES