Journal Home Page

OPEN ACCESS

Artificial Intelligence as Co-Artist: A Phenomenological Study of Human-Machine Collaboration in Fine Arts

Raimoon Kaysar ORCID: https://orcid.org/ Department of Fine Arts in Drawing & Painting Faculty of Fine & Performing Arts Shanto-Mariam University of Creative Technology Dhaka, Bangladesh

Prof. Dr Kazi Abdul Mannan Department of Business Administration Faculty of Business Shanto-Mariam University of Creative Technology Dhaka, Bangladesh Email: drkaziabdulmannan@gmail.com ORCID: https://orcid.org/0000-0002-7123-132X   Corresponding author: Raimoon Kaysar: nabiharahininfinity@gmail.com

Pedagog. res. dev. 2026, 5(2); https://doi.org/10.64907/xkmf.v5i2.prd.3

Submission received: 2 April 2026 / Revised: 20 May 2026 / Accepted: 25 May 2026 / Published: 29 May 2026

Download (PDF)

Abstract

The emergence of artificial intelligence (AI) as a co-creative agent has fundamentally transformed contemporary fine arts, challenging conventional understandings of authorship, creativity, and artistic agency. This study examines human-machine collaboration through a phenomenological lens, focusing on how artists experience and interpret AI as a co-artist. Employing a qualitative research design based on secondary data, including artist statements, exhibition texts, and scholarly discourse, the study identifies key themes such as distributed authorship, dialogic interaction, algorithmic unpredictability, and the reconfiguration of creativity as an emergent process. The findings suggest that AI is no longer perceived as a passive tool but as an active participant that reshapes artistic decision-making and expands creative possibilities. By integrating phenomenology, posthumanism, and distributed cognition, the study offers a comprehensive theoretical framework for understanding AI-assisted art. It further highlights critical tensions between human control and machine autonomy, as well as ethical concerns related to data use and authorship. The research concludes that AI-driven collaboration represents a significant paradigm shift in fine arts, necessitating new conceptual and methodological approaches to artistic practice in the digital age.

Keywords: artificial intelligence, co-creation, phenomenology, human-machine collaboration, distributed creativity, authorship, digital art

1. Introduction

The integration of artificial intelligence (AI) into contemporary artistic practice represents one of the most significant paradigm shifts in the history of fine arts. For centuries, artistic creation has been framed as an inherently human endeavour, grounded in subjective experience, intentionality, and embodied perception. From Renaissance humanism to modernist individualism, the figure of the artist has traditionally been positioned as a singular creative authority whose originality and expressive capacity define the value of the artwork. However, the rapid advancement of AI technologies, particularly machine learning, neural networks, and generative algorithms, has begun to destabilise these assumptions, introducing non-human agents into the creative process in unprecedented ways (Boden, 2016; Manovich, 2019).

Recent developments in AI-driven systems, such as generative adversarial networks (GANs) and transformer-based models, have enabled machines to produce images, texts, and multimedia outputs that exhibit aesthetic coherence and stylistic complexity. These outputs often blur the boundaries between human and machine authorship, raising fundamental questions about creativity, originality, and artistic intention (Elgammal et al., 2017; Goodfellow et al., 2014). In the context of fine arts, AI is no longer confined to the role of a passive tool; rather, it is increasingly conceptualised as an active collaborator or “co-artist” that participates in the generation of meaning and form.

This shift has profound implications for both artistic practice and theoretical discourse. On one hand, artists are experimenting with AI as a generative partner, incorporating algorithmic processes into their workflows and embracing the unpredictability of machine outputs. On the other hand, scholars and critics are grappling with the ontological and epistemological consequences of these practices, particularly in relation to authorship, agency, and the nature of creativity (McCormack et al., 2019). The emergence of AI as a co-artist challenges the anthropocentric foundations of art theory, inviting new perspectives that account for the agency of non-human systems.

Within this evolving landscape, phenomenology offers a valuable framework for understanding the experiential dimensions of human-machine collaboration. Rooted in the philosophical traditions of Edmund Husserl and later developed by Maurice Merleau-Ponty, phenomenology emphasises the study of lived experience and the structures of perception (Merleau-Ponty, 1962). Rather than focusing solely on the technical capabilities of AI, a phenomenological approach seeks to explore how artists experience and interpret their interactions with intelligent systems. This perspective foregrounds the subjective and relational aspects of co-creation, highlighting the ways in which meaning emerges through engagement with both human and non-human agents.

The notion of AI as a co-artist also intersects with broader theoretical developments in posthumanism and new materialism. These perspectives challenge the centrality of the human subject and emphasise the distributed nature of agency across networks of humans, technologies, and environments (Braidotti, 2013; Hayles, 1999). In this context, creativity is no longer understood as an exclusively human attribute but as an emergent property of complex interactions involving multiple actors. This reconceptualisation has significant implications for how artistic value is defined and evaluated, particularly in relation to originality, authenticity, and authorship.

Despite the growing body of literature on AI and art, there remains a need for in-depth qualitative research that examines the lived experiences of artists working with AI systems. Much of the existing scholarship focuses on technical innovation or theoretical speculation, often overlooking the phenomenological dimensions of artistic practice. By analysing secondary data such as artist statements, interviews, and exhibition texts, this study seeks to address this gap, offering a nuanced understanding of how human-machine collaboration is experienced and articulated within fine arts.

The central aim of this research is to explore AI as a co-artist through a phenomenological lens, examining how artists conceptualise and engage with intelligent systems in their creative processes. Specifically, the study addresses the following research questions:

• How do artists perceive and define AI as a co-creative partner?
• What kinds of lived experiences emerge from human-machine collaboration in fine arts?
• How does the presence of AI reshape traditional notions of authorship, creativity, and artistic agency?

By addressing these questions, the study contributes to ongoing debates about the role of AI in creative industries and the future of artistic practice. It also provides a theoretical and methodological framework for analysing human-machine collaboration, drawing on insights from phenomenology, posthumanism, and distributed cognition.

Ultimately, this research argues that the integration of AI into fine arts represents not merely a technological innovation but a fundamental transformation in the ontology of creativity. As artists continue to engage with AI as a co-artist, new forms of expression and interpretation are likely to emerge, challenging established paradigms and opening up new possibilities for artistic exploration. Understanding these developments requires a shift from viewing AI as a tool to recognising it as an active participant in the creative process, one that reshapes both the experience and the meaning of art in the digital age.

2. Literature Review

The intersection of artificial intelligence and artistic creativity has become a focal point of interdisciplinary research, encompassing fields such as computer science, art theory, media studies, and philosophy. Early explorations of computational creativity focused on rule-based systems capable of generating visual patterns or musical compositions. However, the advent of machine learning and deep neural networks has significantly expanded the creative potential of AI, enabling systems to learn from large datasets and produce outputs that exhibit stylistic nuance and variation (Boden, 2016).

One of the most influential developments in this domain is the introduction of generative adversarial networks (GANs), which consist of two neural networks-a generator and a discriminator-that operate in a competitive framework to produce increasingly realistic outputs (Goodfellow et al., 2014). GANs have been widely used in artistic contexts to generate images that mimic established styles or create entirely new visual forms. Elgammal et al. (2017) further advanced this approach through the development of Creative Adversarial Networks (CAN), which aim to deviate from learned styles to produce novel and aesthetically engaging artworks.

Despite these advancements, debates persist regarding the extent to which AI can be considered truly creative. Boden (2016) distinguishes between combinational, exploratory, and transformational creativity, suggesting that while AI systems excel at combining and exploring existing patterns, they may be limited in their ability to achieve genuine transformation. Similarly, Colton et al. (2015) argue that computational creativity should be evaluated not only in terms of output but also in relation to the processes and intentions underlying artistic production.

2.1 Rethinking Authorship and Agency

The rise of AI-generated art has prompted a reexamination of authorship, a concept that has long been central to art historical discourse. Traditional models of authorship emphasise the role of the individual artist as the originator of creative works. However, this model has been increasingly challenged by poststructuralist theories that question the stability and singularity of the authorial voice (Barthes, 1977; Foucault, 1984).

Barthes’ notion of the “death of the author” suggests that meaning is not fixed by the creator but emerges through the interaction between text and reader. Foucault (1984) further complicates this idea by conceptualising the author as a function of discourse rather than a fixed identity. These theoretical perspectives provide a foundation for understanding the distributed nature of authorship in AI-driven art, where creative outputs result from interactions among multiple agents, including programmers, datasets, algorithms, and users.

In this context, agency becomes a critical point of inquiry. Gunkel (2018) argues that as machines become more autonomous, they challenge traditional distinctions between tools and agents. While AI systems do not possess consciousness or intentionality in the human sense, their capacity to generate novel outputs and influence creative decisions suggests a form of functional agency. This has led some scholars to advocate for the recognition of AI as a co-author or collaborator, rather than merely an instrument.

McCormack et al. (2019) emphasise that the creative process in AI-assisted art is inherently collaborative, involving iterative interactions between human and machine. They argue that authorship should be understood as a dynamic and distributed phenomenon, reflecting the contributions of both human and non-human actors. This perspective aligns with broader shifts in contemporary art, where collaboration and interdisciplinarity are increasingly valued.

2.2 Phenomenology and the Experience of Artistic Creation

Phenomenology provides a critical framework for examining the experiential dimensions of artistic practice, particularly in relation to perception, embodiment, and intentionality. Husserl’s foundational work emphasises the importance of returning “to the things themselves,” focusing on how phenomena are experienced rather than how they are objectively defined. Merleau-Ponty (1962) extends this approach by emphasising the embodied nature of perception, arguing that experience is always situated within a relational context.

In the realm of art, phenomenology has been used to explore both the creation and reception of artworks. Artists are understood as engaged in a process of meaning-making that involves interaction with materials, tools, and environments. When AI is introduced into this process, the phenomenological experience becomes more complex, involving interactions with systems that exhibit autonomous or semi-autonomous behaviour.

Recent studies suggest that artists working with AI often describe their experiences in terms of dialogue, collaboration, and co-presence (Candy & Edmonds, 2018). These descriptions highlight the relational nature of human-machine interaction, where meaning emerges through iterative engagement. The unpredictability of AI outputs is frequently cited as a source of inspiration, prompting artists to reconsider their assumptions and explore new creative directions.

2.3 Posthumanism and Distributed Creativity

Posthumanist theory provides a broader philosophical context for understanding the role of AI in artistic practice. By challenging anthropocentric assumptions, posthumanism emphasises the interconnectedness of humans, technologies, and environments (Braidotti, 2013). This perspective is particularly relevant for analysing AI as a co-artist, as it recognises the agency of non-human actors within creative processes.

Hayles (1999) argues that the boundaries between human and machine are increasingly blurred in the digital age, leading to new forms of hybrid subjectivity. In artistic contexts, this hybridity is reflected in the integration of algorithmic processes into creative workflows, resulting in artworks that are co-produced by human and machine.

The concept of distributed creativity further supports this perspective by framing creativity as an emergent property of interactions among multiple agents (Sawyer, 2012). Rather than being located within an individual mind, creativity is understood as a process that unfolds across networks of people, tools, and systems. In AI-assisted art, this network includes not only the artist and the algorithm but also the data used to train the system and the cultural context in which the work is produced.

2.4 Ethical and Critical Perspectives

While the integration of AI into art offers new possibilities, it also raises important ethical and critical concerns. One major issue is the use of training data, which often consists of copyrighted or culturally significant works. This raises questions about intellectual property, appropriation, and consent (Manovich, 2019).

Additionally, biases embedded in datasets can influence the outputs of AI systems, potentially reinforcing existing inequalities and stereotypes. Scholars have called for greater transparency and accountability in the development and use of AI technologies, particularly in creative contexts where cultural representation is at stake (McCormack et al., 2019).

Another concern is the potential commodification of creativity, as AI-generated art becomes increasingly marketable. The commercialisation of AI art may prioritise novelty and efficiency over critical engagement, leading to a homogenisation of artistic expression. These issues underscore the need for a critical framework that balances innovation with ethical responsibility.

3. Theoretical Framework

This study is grounded in an interdisciplinary theoretical framework that integrates phenomenology, posthumanism, and theories of distributed cognition and creativity. These frameworks collectively provide a conceptual foundation for understanding artificial intelligence (AI) not merely as a technological tool but as an active participant in artistic production. By synthesising these perspectives, the study offers a nuanced lens through which human-machine collaboration in fine arts can be critically examined.

3.1 Phenomenology of Co-Creation

Phenomenology, as established by Edmund Husserl and later developed by Maurice Merleau-Ponty, centres on the analysis of lived experience and the structures of consciousness. Husserl’s concept of intentionality-the idea that consciousness is always directed toward something-provides a critical starting point for understanding artistic creation as a process of meaning-making (Husserl, 1970). In traditional artistic practice, intentionality is attributed to the human artist, whose perceptions, emotions, and decisions shape the final artwork.

However, the introduction of AI into the creative process complicates this model. While AI systems do not possess consciousness in a phenomenological sense, they generate outputs that influence human perception and decision-making. This creates what can be described as a hybrid intentional field, where human intentionality interacts with algorithmic processes. From a phenomenological perspective, the focus shifts from the internal states of the machine to the experiential reality of the human artist engaging with AI-generated outputs.

Merleau-Ponty’s (1962) emphasis on embodiment and perception further enriches this analysis. Artistic creation is not a purely cognitive activity but an embodied practice involving interaction with materials, tools, and environments. When AI is integrated into this process, it becomes part of the artist’s perceptual field, shaping how possibilities are perceived and acted upon. The experience of co-creation with AI can thus be understood as a form of extended embodiment, where the boundaries of the self are expanded through interaction with technological systems.

This phenomenological approach allows for an exploration of how artists experience AI as a co-creative presence. Rather than asking whether AI is truly creative, the framework emphasises how creativity is experienced and interpreted in the context of human-machine interaction. This shift from ontological to experiential inquiry is central to the study’s analytical perspective.

3.2 Posthumanism and Decentered Agency

Posthumanism provides a critical theoretical lens for rethinking the role of AI in artistic practice. Emerging from critiques of humanism, posthumanist theory challenges the notion of the human subject as autonomous, rational, and central to all forms of meaning-making (Braidotti, 2013). Instead, it emphasises the interconnectedness of humans, technologies, and environments, advocating for a more relational understanding of agency.

In the context of AI-assisted art, posthumanism enables a reconceptualisation of the artist-machine relationship. Rather than viewing AI as subordinate to human intention, it is understood as part of a network of actors that collectively contribute to creative outcomes. This perspective aligns with Hayles’ (1999) argument that the boundaries between human and machine are increasingly blurred in the digital age, giving rise to hybrid forms of subjectivity.

The concept of decentered agency is particularly relevant here. Agency is no longer located solely within the human subject but is distributed across human and non-human entities. AI systems, while lacking consciousness, exhibit forms of operational autonomy that influence the direction and outcome of creative processes. This challenges traditional hierarchies and invites a rethinking of authorship as a shared or distributed phenomenon.

Posthumanism also raises important ethical and philosophical questions. If AI is considered a co-artist, what are the implications for responsibility, ownership, and accountability? While this study does not aim to resolve these questions, it acknowledges their significance and situates them within the broader theoretical framework.

3.3 Distributed Cognition and Creativity

The theory of distributed cognition further complements the phenomenological and posthumanist perspectives by emphasising the role of external systems in cognitive processes. Originally developed within cognitive science, distributed cognition posits that thinking and problem-solving are not confined to the individual mind but are distributed across interactions with tools, environments, and other agents (Hutchins, 1995).

In artistic contexts, this perspective has been extended to the concept of distributed creativity, which frames creative processes as emergent phenomena arising from interactions among multiple contributors (Sawyer, 2012). This approach is particularly relevant for understanding AI-assisted art, where creativity emerges from the interplay between human intuition, algorithmic generation, and data-driven processes.

AI systems function as cognitive extensions, augmenting the artist’s capacity to generate, evaluate, and refine ideas. For example, generative models can produce a vast array of visual variations, enabling artists to explore possibilities that might not have been conceived independently. This collaborative dynamic aligns with the notion of co-creation, where the final artwork is the result of iterative exchanges between human and machine.

Moreover, distributed creativity challenges the traditional emphasis on originality as an individual achievement. Instead, it highlights the importance of interaction, adaptation, and emergence. In this framework, creativity is not a fixed property but a dynamic process that unfolds over time through engagement with multiple agents and systems.

3.4 Integrative Framework

By integrating phenomenology, posthumanism, and distributed cognition, this study establishes a comprehensive theoretical framework for analysing AI as a co-artist. Phenomenology provides insight into the lived experiences of artists, posthumanism reconfigures notions of agency and subjectivity, and distributed cognition explains the collaborative dynamics of creative processes.

Together, these perspectives enable a holistic understanding of human-machine collaboration in fine arts. They move beyond binary distinctions between human and machine, instead emphasising relationality, interaction, and emergence. This integrative framework serves as the foundation for the methodological approach and analysis presented in the subsequent sections.

4. Methodology

This study adopts a qualitative research design grounded in interpretive and phenomenological traditions. The primary aim is to explore the lived experiences and conceptual interpretations of human-machine collaboration in fine arts, particularly in relation to AI as a co-artist. Given the exploratory and interpretive nature of the research questions, a qualitative approach is most appropriate, as it allows for in-depth analysis of subjective meanings and contextual nuances (Creswell & Poth, 2018).

The study employs a phenomenological orientation, focusing on how artists experience and make sense of their interactions with AI systems. Rather than attempting to measure or quantify creativity, the research seeks to uncover the essence of co-creative experiences as described in textual sources. This approach aligns with the broader objective of understanding AI not merely as a technical system but as a participant in artistic practice.

4.1 Data Collection: Secondary Data Sources

Due to the conceptual and theoretical nature of the study, data are derived from secondary sources. Secondary data analysis is particularly suitable for research that seeks to synthesise existing knowledge and interpret documented experiences across multiple contexts (Johnston, 2017). The use of secondary data also enables access to a diverse range of perspectives, including those of artists, curators, and scholars.

The data sources for this study include:

• Artist statements and interviews: These provide firsthand accounts of creative processes and experiences with AI.
• Exhibition catalogues and curatorial essays: These contextualise artworks within broader artistic and theoretical frameworks.
• Scholarly articles and books: These offer critical analyses and conceptual discussions of AI and art.
• Case studies of AI-assisted artworks: These document specific instances of human-machine collaboration, providing concrete examples for analysis.

These sources were selected based on their relevance to the research questions and their contribution to understanding AI as a co-creative agent. Emphasis was placed on materials that explicitly address the experiential and interpretive dimensions of artistic practice.

4.2 Sampling Strategy

A purposive sampling strategy was employed to identify relevant data sources. Purposive sampling involves selecting materials that are particularly informative and aligned with the research objectives (Patton, 2015). In this study, sources were chosen based on the following criteria:

• Direct engagement with AI in artistic practice
• Availability of descriptive or reflective content
• Relevance to themes of authorship, creativity, and collaboration

This approach ensures that the data is both meaningful and contextually rich, enabling a detailed exploration of the phenomena under investigation.

4.3 Data Analysis

The analysis follows a thematic and phenomenological approach, combining systematic coding with interpretive reflection. The process consists of several stages:

• Data Familiarisation: All selected materials were carefully read and reviewed to gain an overall understanding of the content. This initial stage involved identifying key concepts, recurring themes, and significant statements related to human-machine collaboration.
• Coding and Categorisation: Relevant segments of text were coded based on their thematic content. Codes were developed inductively, allowing themes to emerge from the data rather than being imposed. Examples of initial codes include “AI as collaborator,” “loss of control,” “unexpected outputs,” and “dialogic interaction.”
• Theme Development: Codes were grouped into broader themes that capture the essential aspects of the phenomenon. These themes were refined through iterative analysis, ensuring that they accurately reflect the data and align with the research questions.
• Phenomenological Interpretation: The final stage involved interpreting the themes through a phenomenological lens, focusing on the lived experiences and meanings associated with AI-assisted creation. This process included identifying the essence of co-creative experiences and exploring how they are articulated by artists.

Phenomenological reduction, or epoché, was applied to minimise researcher bias and bracket preconceived assumptions (Moustakas, 1994). While complete objectivity is not achievable, this approach enhances the rigour and credibility of the analysis.

4.4 Trustworthiness and Rigour

To ensure the quality and credibility of the research, several strategies were employed:

• Credibility: Achieved through the use of diverse and authoritative sources, as well as careful interpretation of data.
• Transferability: Enhanced by providing detailed descriptions of context and methodology, allowing readers to assess applicability to other settings.
• Dependability: Maintained through a transparent and systematic analytical process.
• Confirmability: Supported by grounding interpretations in documented evidence rather than subjective opinion (Lincoln & Guba, 1985).

4.5 Ethical Considerations

As the study relies exclusively on secondary data, it does not involve direct interaction with human participants. However, ethical considerations remain important. All sources were properly cited in accordance with APA (7th ed.) guidelines, ensuring academic integrity and respect for intellectual property.

Additionally, care was taken to represent the perspectives of artists and scholars accurately, without misinterpretation or distortion. The study also acknowledges the broader ethical implications of AI in art, including issues related to data usage, bias, and authorship (Mannan & Farhana, 2026).

4.6 Limitations

While the use of secondary data provides valuable insights, it also presents certain limitations. The analysis is dependent on the availability and quality of existing sources, which may vary in depth and perspective. Furthermore, the absence of primary data means that the study cannot capture real-time experiences or probe deeper into specific issues.

Despite these limitations, the methodological approach is well-suited to the exploratory nature of the research and provides a robust foundation for understanding AI as a co-artist.

5. Findings and Analysis

The analysis of secondary data, including artist statements, interviews, exhibition texts, and scholarly discussions, reveals a complex and evolving understanding of artificial intelligence (AI) as a co-artist in contemporary fine arts. Through thematic coding and phenomenological interpretation, several key patterns emerge that illuminate how artists experience, conceptualise, and negotiate human-machine collaboration. These findings are organised into five major themes: AI as a creative partner, dialogic interaction and co-presence, negotiation of control and autonomy, reconfiguration of creativity, and materiality, data, and the aesthetics of algorithmic production.

5.1 AI as a Creative Partner

A central finding across the data is the consistent reframing of AI from a passive tool to an active creative partner. Artists frequently describe AI systems using relational language such as “collaborator,” “co-author,” or “creative companion,” suggesting a shift in how agency is attributed within the artistic process. This aligns with broader discussions in computational creativity, where AI systems are recognised for their capacity to generate novel outputs that influence human decision-making (Colton et al., 2015; McCormack et al., 2019).

Unlike traditional tools, which function as extensions of human intention, AI systems introduce elements of unpredictability and generativity. For instance, artists working with generative adversarial networks (GANs) often emphasise the system’s ability to produce unexpected visual forms that challenge preconceived ideas. These outputs are not merely executed commands but are perceived as contributions that shape the direction of the artwork (Elgammal et al., 2017).

From a phenomenological perspective, this experience can be understood as a reorientation of intentionality. While the artist initiates the process, the AI system introduces variations that redirect attention and influence subsequent choices. The resulting artwork emerges from a reciprocal dynamic, where both human and machine contribute to the unfolding of form and meaning. This supports the notion of distributed creativity, in which creative agency is shared across multiple actors and systems (Sawyer, 2012).

5.2 Dialogic Interaction and Co-Presence

Another prominent theme is the experience of dialogic interaction between the artist and the AI system. Artists frequently describe their engagement with AI as a form of conversation or exchange, where outputs generated by the system are interpreted, evaluated, and responded to in an iterative process. This dialogic structure resembles what Candy and Edmonds (2018) describe as practice-based interaction, where meaning emerges through ongoing engagement rather than predetermined outcomes.

The sense of co-presence-feeling as though one is interacting with another agent-is particularly significant. Although AI lacks consciousness, its outputs are often perceived as responsive or intentional. This perception is not based on the machine’s internal state but on the phenomenological experience of interaction. As Merleau-Ponty (1962) suggests, perception is inherently relational, shaped by the interplay between subject and environment. In this context, AI becomes part of the artist’s perceptual field, contributing to the experience of co-creation.

This dialogic engagement also introduces a temporal dimension to the creative process. Rather than following a linear trajectory, the process unfolds through cycles of generation, interpretation, and modification. Each iteration introduces new possibilities, prompting the artist to reconsider their intentions and explore alternative directions. This iterative structure reflects the dynamic nature of human-machine collaboration and highlights the importance of responsiveness and adaptability.

5.3 Negotiating Control and Autonomy

The relationship between control and autonomy emerges as a critical tension in AI-assisted art. Artists often navigate a delicate balance between guiding the creative process and allowing the AI system to operate independently. This tension is evident in the ways artists describe their interactions with algorithmic systems, oscillating between control-oriented strategies and openness to machine-generated variation.

On one hand, artists exercise control by selecting training data, adjusting parameters, and curating outputs. These actions reflect a desire to maintain authorship and ensure that the final work aligns with their artistic vision. On the other hand, many artists deliberately relinquish control at certain stages, embracing the unpredictability of AI as a source of inspiration. This duality reflects what McCormack et al. (2019) identify as a hybrid creative process, where human intention and machine autonomy coexist.

From a theoretical standpoint, this tension can be interpreted through the lens of posthumanism. The decentering of human agency challenges traditional hierarchies, positioning AI as an active participant rather than a subordinate tool (Braidotti, 2013). However, this shift does not eliminate the role of the human artist; instead, it reconfigures it. The artist becomes a mediator or facilitator, orchestrating interactions between themselves and the machine.

The negotiation of control also raises questions about responsibility and authorship. If an AI system generates an unexpected or controversial output, who is accountable? While this study does not provide definitive answers, the findings suggest that artists are increasingly aware of these complexities and actively engage with them in their practice.

5.4 Reconfiguration of Creativity

The integration of AI into artistic practice necessitates a redefinition of creativity itself. Traditional models of creativity emphasise originality, intentionality, and individual expression. However, the findings indicate a shift toward understanding creativity as an emergent and relational process.

Artists often describe their role not as sole creators but as curators, editors, or interpreters of machine-generated content. This shift reflects a move away from the idea of creativity as a product of individual genius toward a more collaborative and process-oriented conception. As Boden (2016) notes, AI systems are particularly effective at exploratory creativity, generating variations within a defined conceptual space. When combined with human interpretation, these variations can lead to novel and meaningful outcomes.

The concept of originality is also reexamined in this context. AI-generated outputs are derived from training data, which consists of existing images or styles. This raises questions about whether such outputs can be considered truly original. However, many artists argue that originality lies not in the source material but in the process of selection, transformation, and contextualization. This perspective aligns with poststructuralist theories that view meaning as contingent and relational rather than fixed (Barthes, 1977).

5.5 Materiality, Data, and Algorithmic Aesthetics

A final theme concerns the material and aesthetic dimensions of AI-assisted art. While digital technologies are often associated with immateriality, the findings highlight the importance of data and algorithms as new forms of artistic material. Training datasets, model architectures, and computational processes all contribute to the final aesthetic outcome.

Artists working with AI frequently engage with the materiality of data, exploring how different datasets produce distinct visual characteristics. For example, training a model on historical paintings may result in outputs that reflect classical styles, while using contemporary images may yield more experimental forms. This interaction between data and aesthetics underscores the role of AI as both a medium and a collaborator.

Moreover, the aesthetics of AI-generated art often emphasise features such as repetition, distortion, and hybridity. These characteristics reflect the underlying computational processes and distinguish AI-assisted works from traditional forms of art. As Manovich (2019) argues, AI aesthetics are shaped by the logic of algorithms, which introduce new visual languages and modes of representation.

6. Discussion

The findings of this study provide a comprehensive understanding of how artificial intelligence (AI) functions as a co-artist within contemporary fine arts, revealing significant implications for theory, practice, and critical discourse. By interpreting these findings through the integrated lenses of phenomenology, posthumanism, and distributed cognition, this discussion situates AI-assisted creativity within broader epistemological and ontological transformations.

6.1 Reframing Artistic Agency and Authorship

One of the most significant implications of this study is the reconfiguration of artistic agency. Traditional models position the artist as the primary locus of creativity, responsible for both the conception and execution of the artwork. However, the findings demonstrate that in AI-assisted practices, agency is distributed across human and non-human actors, including algorithms, datasets, and computational infrastructures.

This shift aligns with posthumanist critiques of anthropocentrism, which challenge the assumption that humans are the sole agents of meaning-making (Braidotti, 2013). By recognising AI as a co-artist, contemporary practice moves toward a more relational understanding of agency, where creative outcomes emerge from interactions rather than individual intention.

Authorship, consequently, becomes a fluid and contested concept. The idea of a singular author is replaced by a networked model of co-authorship, where multiple contributors shape the final work. This resonates with Foucault’s (1984) notion of the “author function,” which conceptualises authorship as a construct rather than an inherent property. In AI-assisted art, this construct becomes even more complex, as it must account for non-human contributions.

6.2 Phenomenological Implications: Experience and Perception

From a phenomenological perspective, the experience of co-creating with AI introduces new modes of perception and engagement. Artists describe a sense of dialogue and co-presence, suggesting that AI systems are integrated into their perceptual and cognitive processes. This integration can be understood as an extension of embodiment, where technological systems become part of the artist’s experiential world (Merleau-Ponty, 1962).

The unpredictability of AI outputs plays a crucial role in shaping this experience. Rather than following a predetermined plan, artists respond to emergent possibilities generated by the system. This responsiveness reflects a shift from intentional control to adaptive engagement, where creativity arises through interaction.

Phenomenology also highlights the importance of temporality in AI-assisted creation. The iterative nature of human-machine collaboration introduces a dynamic temporal structure, characterised by cycles of generation and interpretation. This temporal dimension reinforces the idea of creativity as a process rather than a static outcome.

6.3 Distributed Creativity and Cognitive Extension

The findings strongly support the concept of distributed creativity, which posits that creative processes are not confined to individual minds but emerge from interactions among multiple agents and systems (Sawyer, 2012). AI systems function as cognitive extensions, augmenting the artist’s capacity to explore, generate, and evaluate ideas.

This perspective challenges traditional distinctions between internal and external cognition. As Hutchins (1995) argues, cognitive processes are often distributed across tools and environments. In the context of AI-assisted art, the algorithm becomes part of the cognitive system, contributing to the generation of ideas and the evaluation of outcomes.

The notion of cognitive extension also raises important questions about skill and expertise. If AI systems can generate complex visual forms, what role does human skill play? The findings suggest that expertise shifts from manual execution to conceptual and interpretive abilities. Artists must develop new competencies, including an understanding of algorithmic processes and the ability to critically engage with machine-generated outputs.

6.4 Ethical and Critical Considerations

The integration of AI into artistic practice raises several ethical and critical issues that warrant careful consideration. One key concern is the use of training data, which often includes copyrighted or culturally significant materials. This raises questions about intellectual property, authorship, and appropriation (Manovich, 2019).

Bias in datasets is another critical issue. AI systems learn from existing data, which may reflect social and cultural biases. As a result, the outputs of these systems can perpetuate or amplify these biases, influencing artistic representation. Addressing this issue requires greater transparency and accountability in the development and use of AI technologies (McCormack et al., 2019).

The commercialisation of AI-generated art also presents challenges. As AI tools become more accessible, there is a risk that artistic production may be driven by efficiency and market demand rather than critical engagement. This could lead to a homogenization of aesthetic forms and a reduction in diversity.

6.5 Toward a New Ontology of Art

Ultimately, the findings of this study point toward a new ontology of art, one that acknowledges the role of non-human agents in creative processes. AI-assisted art challenges traditional boundaries between subject and object, creator and tool, human and machine. In doing so, it opens up new possibilities for artistic expression and theoretical inquiry.

This new ontology is characterised by relationality, emergence, and hybridity. Art is no longer seen as the product of isolated individuals but as the outcome of complex interactions involving multiple agents and systems. This perspective requires a rethinking of key concepts such as creativity, authorship, and originality, as well as the development of new critical frameworks.

7. Conclusion

This study has explored the role of artificial intelligence (AI) as a co-artist within contemporary fine arts, offering a phenomenological analysis of human-machine collaboration grounded in qualitative secondary data. The findings demonstrate that AI is increasingly understood not as a passive instrument but as an active and generative participant in artistic processes. This shift challenges long-standing assumptions about creativity as an exclusively human capacity and repositions artistic production as a relational and distributed phenomenon.

Through the integration of phenomenology, posthumanism, and distributed cognition, the study provides a comprehensive framework for interpreting the evolving dynamics of artistic practice. Phenomenology reveals how artists experience AI as a co-creative presence, characterised by dialogic interaction, unpredictability, and co-presence. Posthumanist perspectives further destabilise anthropocentric models of authorship, emphasising the role of non-human agents in shaping creative outcomes. Meanwhile, the concept of distributed creativity underscores the emergent nature of artistic production, arising from interactions among human intention, algorithmic processes, and data infrastructures.

The study also highlights significant tensions inherent in AI-assisted art, particularly between control and autonomy. Artists navigate a complex balance between guiding the creative process and allowing AI systems to generate unexpected outputs. This negotiation redefines the role of the artist as a mediator, curator, and interpreter rather than a sole creator. Additionally, ethical concerns related to data usage, algorithmic bias, and intellectual property remain critical challenges that require ongoing attention.

Importantly, the findings suggest that AI-driven collaboration is not merely a technological innovation but a fundamental transformation in the ontology of art. As creative practices continue to incorporate intelligent systems, traditional categories such as originality, authorship, and artistic intention must be reconsidered. This transformation calls for new critical frameworks and methodological approaches capable of addressing the complexities of human-machine interaction.

Future research should extend this inquiry by incorporating primary empirical studies, including interviews and ethnographic observations of artists working with AI. Such approaches would deepen understanding of lived experiences and provide further insight into the evolving relationship between humans and intelligent systems. Ultimately, recognising AI as a co-artist opens new possibilities for artistic exploration while challenging the boundaries of creativity in the digital era.

References

Barthes, R. (1977). Image, music, text. Hill and Wang.

Boden, M. A. (2016). AI: Its nature and future. Oxford University Press.

Braidotti, R. (2013). The posthuman. Polity Press.

Candy, L., & Edmonds, E. (2018). Practice-based research in the creative arts: Foundations and futures. Leonardo, 51(1), 63-69.

Colton, S., Wiggins, G. A., & Pease, A. (2015). Computational creativity theory: The FACE and IDEA models. Proceedings of the International Conference on Computational Creativity, 90-97.

Creswell, J. W., & Poth, C. N. (2018). Qualitative inquiry and research design: Choosing among five approaches (4th ed.). Sage.

Elgammal, A., Liu, B., Elhoseiny, M., & Mazzone, M. (2017). CAN: Creative adversarial networks. Proceedings of the International Conference on Computational Creativity, 96-103.

Foucault, M. (1984). What is an author? In P. Rabinow (Ed.), The Foucault reader (pp. 101-120). Pantheon Books.

Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2014). Generative adversarial nets. Advances in Neural Information Processing Systems, 27, 2672-2680.

Hayles, N. K. (1999). How we became posthuman. University of Chicago Press.

Husserl, E. (1970). The crisis of European sciences and transcendental phenomenology. Northwestern University Press.

Hutchins, E. (1995). Cognition in the wild. MIT Press.

Johnston, M. P. (2017). Secondary data analysis: A method whose time has come. Qualitative and Quantitative Methods in Libraries, 3(3), 619-626.

Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Sage.

Mannan, K.A., & Farhana, K.M. (2026). The Principles of Qur’anic Research Methodology: Deriving the Process of Knowledge from Revelation. KMF Publishers. Open Access (CC BY 4.0). DOI: https://doi.org/10.64907/xkmf.book.pqrm.26.02.12

Manovich, L. (2019). AI aesthetics. Strelka Press.

McCormack, J., Gifford, T., & Hutchings, P. (2019). Autonomy, authenticity, authorship and intention in computer-generated art. Digital Creativity, 30(4), 267-282.

Merleau-Ponty, M. (1962). Phenomenology of perception. Routledge.

Moustakas, C. (1994). Phenomenological research methods. Sage.

Patton, M. Q. (2015). Qualitative research and evaluation methods (4th ed.). Sage.

Sawyer, R. K. (2012). Explaining creativity: The science of human innovation (2nd ed.). Oxford University Press.