SME employee–GenAI collaborative creativity (EGCC): translating creative affordances into social impact Open Access

Creativity is a crucial source of innovation development and represents a competitive advantage for small and medium-sized enterprises (SMEs) to compete with large companies (Castillo-Vergara et al., 2021). However, employees in SMEs often face barriers, such as high workloads and limited budgets (Castillo-Vergara et al., 2021), making experimentation and engagement in creative tasks difficult. SME employees often lack either formal creative or innovation roles or structured processes and support (Castillo-Vergara et al., 2021), suggesting that creativity is often an individual responsibility. Employee proactivity in innovation is essential (Madihalli et al., 2023). Integrating generative artificial intelligence (GenAI) can support SME employees’ creative processes to drive innovation (Rajaram and Tinguely, 2024). GenAI enables SMEs to quickly adapt to market needs, providing a competitive advantage, especially when resources are constrained (Rajaram and Tinguely, 2024). Recent evidence indicates that 43% of SMEs utilize GenAI for various tasks, including brainstorming, language translation, document summarization, research and communication creation; however, only 36% view this technological advancement as beneficial for their workplaces (Inside Small Business, 2024). These perceptions create a paradox. That is, while GenAI can support creative processes and decision-making in SMEs, GenAI value often depends on the balance between human and AI creativity. A balanced approach reflects the complementary integration of human judgment and GenAI capabilities, with value co-created through interactions. The approach fosters creativity and ensures meaningful work and equitable human contributions in SMEs. An imbalance in collaboration occurs when there is an uncritical reliance on GenAI outputs without sufficient human judgment, leading to a loss of autonomy and decision-making capabilities (Krakowski, 2025). To capture this dynamic, this paper develops the concept of SME Employees-GenAI Collaborative Creativity (EGCC).

Beyond firm outcomes, we propose that EGCC is a mechanism through which SMEs can generate social impact. Russell-Bennett and Reid (2026) define social impact as the creation of value for people through business activities that result in improvements in human, economic and environmental phenomena over time. SMEs are active agents driving social impact by embedding socially responsible creative practices into everyday interactions between SME employees and GenAI systems. EGCC is a pathway to socially responsible creative outcomes, through which SMEs can realize value for employees and broader stakeholders to achieve social impact (Russell-Bennett and Reid, 2026). At the micro level, EGCC can help enhance SME employee inclusion and meaningful work, upskilling their digital capabilities through balanced human-AI collaboration. At the meso level, EGCC can support SMEs in adopting responsible, innovative practices that transform how services are designed and delivered, thereby improving well-being, access and participation among different stakeholders. Through EGCC, SME employees can generate more ethically and socially responsible outputs. When implemented, these outputs improve access, fairness and transparency for customers’ experiences and deliverables. For SME employees, this process should translate to meaningful work, equitable participation and enhanced digital capabilities. At the organizational and community level, EGCC shapes more equitable and accessible business ecosystems. This supports SMEs to achieve Sustainable Development Goals 8 (Decent Work and Economic Growth) and 9 (Industry, Innovation and Infrastructure) by enhancing meaningful work and digital capabilities, while also enabling inclusive participation and responsible innovation via employee-GenAI collaboration.

Current studies investigate how employees adopt GenAI and the interplay between human agency and technological capabilities (Brynjolfsson et al., 2025; Creely and Blannin, 2025; Rajaram and Tinguely, 2024). These studies tend to emphasize the productivity and efficiency gains from using GenAI, with less attention to the implications for creativity, particularly how employees and GenAI jointly produce creative outputs through cognitive and social processes (e.g. He et al., 2025; Rajaram and Tinguely, 2024). Although research into the creative outcomes enabled by GenAI is on the increase (e.g. Doshi and Hauser, 2024; Rajaram and Tinguely, 2024), current theoretical frameworks tend to focus on the technology or human implications in isolation, overlooking how SME employees’ interactions with GenAI systems enable creativity. Furthermore, current studies often overlook systemic and institutional influences that either help or constrain creative collaboration. Consequently, there remains a limited integrated, multi-level understanding of how individual, task and contextual conditions jointly shape the enactment of GenAI-enabled creativity.

Our paper seeks to answer the following research questions:

And:

In this paper, creative affordances refer to the action potential of GenAI technology to support idea generation, exploration and elaboration within human-AI interaction (Grilli and Pedota, 2024; O'Toole and Horvát, 2024). Socially responsible creative outcomes refer to AI-assisted outputs that are novel and socially acceptable (Runco and Jaeger, 2012) that fulfil ethical, inclusive and accountability standards (Cheng et al., 2021). These outcomes represent mechanisms for creating value for SME employees, businesses and the wider business ecosystem. By linking affordance enactment with socially responsible creative outcomes, this paper highlights the need to view GenAI as more than a productivity tool but as a technology that shapes responsible and meaningful work practices. Such outcomes are more likely with EGCC because GenAI enables the exploration of possibilities, while human judgment supports refining outputs in line with organizational and societal norms.

We apply an integrative theory synthesis approach to make three contributions. First, we build on affordance theory (Gibson, 1977) to develop the concept of EGCC. The affordance theory lens shifts the focus from a deterministic view of technology as the source of creativity to understanding how GenAI creates affordances that enable creative interactions (DeSchryver et al., 2025). The lens suggests that employees must perceive, interpret and act on the affordances of GenAI in SMEs. Thus, we reconceptualize GenAI as an enabler of creative affordances, with creative outcomes dependent on human enactment. Second, we propose a theoretical framework comprising:

• SME Employee Agency;

• SME GenAI Competency;

• Domain Knowledge;

• Task Compatibility and Reception; and

• Socio-technical and Institutional Conditions.

SMEs often face limited resources and less formalized structures (Leso et al., 2023). The multi-level conceptual approach recognizes these structural realities and integrates individual, task and contextual factors to explain how creative affordances are enacted within SMEs. The framework offers a pathway that links EGCC to socially responsible creative outcomes that can lead to social impact. Third, we introduce three practice-grounded scenarios and provide a practical tip sheet to translate EGCC into actionable strategies for SME managers. These resources help SMEs to integrate GenAI to achieve responsible innovation and translate their creative practices into measurable social impact for SME employees, businesses and ecosystems.

Originating in ecological psychology, the affordance theory (Gibson, 1977) posits that a goal-directed actor perceives an object in the environment as affording the means to achieve that goal. Leonardi (2011) further developed affordance theory to demonstrate that organizational change arises from the dynamic alignment between human agency and technological capabilities. When individuals perceive technology as a limitation, they tend to modify it; conversely, when they recognize technological affordances, they adjust their work habits accordingly. The match of human intentions and technological functionalities that drives innovation. Madihalli et al. (2023) propose the TACT framework to highlight that technology provides affordances and constraints. Technology affordances refer to the actions individuals can perform with a given technology. Constraints refer to the ways in which technology impedes individuals from meeting their goals. The authors caution that people and organizations often fail to recognize a technology’s full potential. For employees in SMEs, GenAI may facilitate creativity by enabling features or functionality, but the technology may also create constraints through blind application. SME employees’ perceptions and interpretation of the GenAI tool, as well as its output, are essential to actualizing the GenAI affordances and achieving creativity.

GenAI presents affordances by democratizing access to creative tools, enabling employees, particularly less experienced creators, to produce output (Doshi and Hauser, 2024). The efficiency of tasks that facilitate ideation also enables employees to generate a broader range of ideas in less time (Medeiros et al., 2025). GenAI also possesses a unique combination of capabilities, including task delegation, user journey mapping, idea generation and concept prototyping (Bilgram and Laarmann, 2023). Finally, GenAI automates routine tasks, allowing employees to focus on higher-level, creative tasks (Madihalli et al., 2023). GenAI offers creative affordances by reducing barriers to experimentation, combining diverse inputs, and accelerating iterative feedback. These affordances can significantly enhance SME employees’ ability to innovate, particularly in the face of resource constraints (Krakowski, 2025). GenAI may also introduce constraints by standardizing the generated content type and producing generic output that lacks distinctiveness (Doshi and Hauser, 2024). On a personal level, GenAI may reduce employees’ convictions of their own creativity and overreliance on technology (Faiella et al., 2025). SMEs rely heavily on their employees’ initiative to generate new ideas (Kmieciak et al., 2012) and often lack the organizational mechanisms available in large organizations to facilitate oversight of technology and outputs. We propose that SMEs can transform limited resources into more effective creative outputs when their employees can actualize creative affordances and mitigate constraints through EGCC. Importantly, whether SME employees can actualize these affordances depends on their capabilities and work context. In particular, SME employees’ domain knowledge shapes how they interpret and recombine AI-generated outputs, while their competence with GenAI tools influences their ability to refine outputs effectively. Furthermore, the nature of the task (e.g. complexity and AI compatibility) may determine which affordances are actionable. These conditions shape how SME employees recognize and enact creative affordances.

Employee creativity can be understood as a multidimensional process, encompassing the behaviours people engage in when generating ideas and the evaluative judgment of the ideas that emerge as novel and useful (Amabile, 1996). Empirical work often focuses on the second definition of creating outputs for novelty and usefulness (Gibson, 1977). For SMEs, the creative-performance behaviours of problem framing, retrieving information, creating ideas and validation (Medeiros et al., 2025) are crucial, as they determine whether promising concepts will progress or stall. These behaviours form a continuum of creativity that runs from minor refinements to radical breakthroughs (Zhou and Hoever, 2014). The employee often influences creative behaviours as an individual, within the work setting, or in the wider social system. Actor-centered models emphasize individual traits, while context-centered models focus on environmental factors such as leadership style, social networks or team characteristics. Interactionist models integrate both, showing that creativity results from the interplay between personal characteristics and organizational factors (Zhou and Hoever, 2014).

Creativity and innovation are shaped by how knowledge is generated and by the nature of the collective, including cognitive, social and organizational structures, and play a critical role in human-to-human creative processes (Acar et al., 2024). In this regard, creativity is not isolated to individual drivers but emerges through interpersonal interactions. Shamout et al. (2025) find that the use of collaborative technologies enhances knowledge sharing and innovative behaviour, reinforcing that these effects are realized through human-to-human engagement alongside technology. Shin et al. (2012) reported that cognitive team diversity positively relates to individual team member creativity when creative self-efficacy is high. Furthermore, transformational leadership moderates this relationship such that team diversity increases creativity most when leadership is strong. In SEM contexts, Ameen et al. (2022) show that organizational creativity influences innovation performance but note that creativity is insufficient without SME resources to act on market knowledge. Madihalli et al. (2023) report that psychological resources, including organizational-based self-esteem and psychological capital, are important to enable creativity among SME employees, while Olsen et al. (2025) showed that employee creativity is shaped by an interplay of job demands (hindrance and challenge), job resources (autonomy) and personal resources (self-efficacy, growth mindset and resilience). He et al. (2025) reinforce that humble leadership directly and positively drives employee creativity, and the relationship is mediated by employee vitality and peer support in SMEs. Overall, these studies underscore the relational and human-to-human nature of creativity in SMEs, emerging from the interplay of individual capacities, interpersonal dynamics and organizational conditions. Regarding human creativity, SME employees’ domain knowledge and cognitive resources are central to the generation, interpretation and development of ideas, reinforcing the role of knowledge along with contextual factors as drivers of creative processes.

GenAI displays computational creativity and reflects what Boden (2009) terms as combinational creativity. GenAI provides employees opportunities for idea generation, synthesis and representation that enhance the human creative process (Doshi and Hauser, 2024; Medeiros et al., 2025). From an affordance perspective, such opportunities represent perceived action possibilities for users to extend their cognitive and imaginative capacities (Leonardi, 2011). Current GenAI models, including large language and image-generation models, do not produce meaning on their own; instead, they respond to human prompts and directions. Modern GenAI generates novel content across multiple modalities, including text, images and code, and is trained on large data sets. For instance, GPT3 massively increases the language model’s parameter count (to 175 billion) to improve its ability to learn new tasks from textual examples (Brown et al., 2020). These foundation models employ neural network architectures that identify patterns within pre-trained data and can generate responses across modalities quickly (Vinchon et al., 2023). Storey et al. (2025) adopted a socio-technical systems perspective to argue that GenAI differs by its ability to generate novel outputs and exhibit emergent behaviours that are difficult to predict or control. Large Language Models (LLMs) focus on processing textual information and generating responses based on probabilities rather than fixed answers (Runco and Jaeger, 2012). Neural probabilistic language models create a unique vector for each word and represent a sentence as a series of these vectors. This allows a single training example to predict similar unseen sentences (Bengio et al., 2003), thereby enabling the generation of novel content. GenAI uses controlled stochasticity to generate varied, naturally sounding output, striking a balance between model confidence and creative diversity (Hettrich et al., 2025). Through human interpretation and interaction, GenAI enables new possibilities by recombining and extending existing knowledge (Ameen et al., 2022). For SMEs, these GenAI affordances generate creative value by offering speed and diversity in idea generation, which supports employees to develop creative outcomes.

Recent studies reinforce the idea that GenAI creativity is co-dependent on human involvement. Ameen et al. (2022) suggest a theoretical conceptualization that underpins how AI might impact creativity within marketing frameworks. The study differentiates between human creativity, human-aided (AI) creativity and AI-driven creativity. The authors explain that AI-aided creativity does not lead to automated creativity but can be a creative co-creator when there is dynamic interaction between humans and machines. Likewise, Yang and Xu (2025) empirically demonstrate that creativity in AI is perceived rather than objective and is shaped by how users perceive the AI to be original and credible. They state that the value of the creative product is a dynamic process. Their findings highlight that creativity is grounded in human interpretation, aligning with the affordance perspective. Amankwah-Amoah et al. (2024) propose that GenAI serves as a collaborative catalyst for human creativity in creative disciplines, raising concerns about authenticity and ethics. The authors call for a balance between automation and the human touch. Chamakiotis and Panteli (2024) view GenAI as generative actors reshaping the epistemic process through both their outputs and their embeddedness in socio-technical practices. These studies suggest that creativity is a relational outcome of human–AI collaboration, achieved when SME employees leverage GenAI’s creative affordances through expertise and responsible judgment. Importantly, this highlights that SME employees’ domain knowledge and GenAI competence will likely function as critical enablers of affordance actualization, shaping how effectively employees can translate AI-generated outputs into meaningful creative outcomes.

Building on Jia et al. (2024) definition and affordances theory (Gibson, 1977), we define SME EGCC as a dynamic co-creative process through which SME employees and GenAI systems jointly produce novel and socially responsible outcomes. Affordance Theory posits that creativity is not inherent in technology but arises when users perceive, interpret and act on the opportunities the technology offers within their specific context (Leonardi, 2011). Recent research on collaborative creativity has revealed that novel ideas often emerge from an iterative conversation between humans and AI, in which the system proposes alternatives and humans refine them into meaningful solutions (Frich et al., 2024). This interaction illustrates the trade-offs between novelty and relevance, as GenAI offers affordances for creative possibilities by accelerating ideation and expanding human cognitive and imaginative capacities (Rajaram and Tinguely, 2024). However, SME employees’ agency and GenAI competence ensure that the outputs meet integrity standards. Wu et al. (2021) outline a Human–AI Co-Creation Model to illustrate how humans and artificial intelligence coexist and co-create by leveraging their complementary strengths. The model comprises six interrelated phases: perceiving, thinking, expressing, collaborating, building and testing, which describe how AI can augment human cognition for enhanced creativity.

GenAI tools help educated professionals to become more productive, efficient and enjoy work (Noy and Zhang, 2023). Doshi and Hauser (2024) found that the use of GenAI increases individual creativity at the expense of collective novelty. Jia et al. (2024) demonstrate that AI assistance through a sequential division of labour increases employee creativity, as the AI manages repetitive work, and employees focus on higher-level problem-solving. However, this effect is skills-biased and tends to benefit higher-skilled employees more, as they possess domain expertise that enables them to convert their cognitive resources into creative outputs. This suggests that employees will not uniformly receive GenAI creative affordances, but they are contingent on individuals’ capacities. The EGCC framework emphasizes that SME employee creativity does not reflect the sum of individual GenAI interactions. Rather, EGCC is based on the premise that creativity is a social and multilevel phenomenon shaped by individuals, task design and institutional conditions, which alter how SME employees enact GenAI creative affordances to support socially responsible creative outcomes.

Bouschery et al. (2023) demonstrate that language models support key innovation tasks, such as text summarization, sentiment analysis, customer insight generation and ideation, thereby improving efficiency and creativity. The authors argue that innovation will rely on hybrid intelligence, where human and AI capabilities complement each other. The model conceptualizes GenAI as a co-creative partner within two stages of ideation and implementation. Hettrich et al. (2025) show GenAI improves project planning performance for novices and professionals. The technology bridges the expertise gap, elevating novices to a level comparable to that of unassisted professionals. The authors position GenAI as an augmentation tool and highlight the need to manage overreliance. Rafner et al. (2025) report that creative agency in human-AI collaboration is dynamic and shifts in response to the user’s stage in the process (e.g. ideation vs refinement), their confidence in steering the AI, and the system’s capacity to accommodate those shifts. At times, participants could not co-create effectively with the AI when the output deviated from users’ intentions and could not be adequately refined, disrupting the flow of creativity. McGuire et al. (2024) reinforced the idea that humans as co-creators have to preserve their creative agency, with technology acting as a responsive collaborator rather than assuming dominance, helping restore creative self-agency for workers.

Optimal EGCC is likely realized when GenAI expands the solution space and employees provide contextual knowledge and ethical guardrails (Vinchon et al., 2023). This mutual relationship encourages originality and requires transparency and fair attribution of both human and machine inputs, thereby reducing plagiarism and preserving human motivation. Reciprocal involvement also fosters mutual agency, which broadens the horizons of creativity and knowledge (Creely and Blannin, 2025). Therefore, EGCC is realized when the affordances of GenAI are actualized through human agency, in which technology amplifies human creative potential and employees exercise critical and ethical control (Creely and Blannin, 2025). We propose that optimal EGCC is contingent on a set of interrelated drivers that shape the actualization of creative affordances. SME Employee Agency and GenAI Competence are micro-level drivers determining whether SME employees can perceive and act on GenAI affordances. Specialized Domain Knowledge and Task Compatibility and Reception represent task-level factors that align GenAI affordances with the work context. Finally, socio-technical and institutional conditions systematically reflect how organizational conditions and broader narratives enable or constrain SME employees’ use of GenAI for creativity. Together, these drivers provide the theoretical foundation for understanding how EGCC enables socially responsible creative outcomes that are mechanisms for social impact.

We build on affordance theory to propose the SME EGCC framework, identifying five key drivers derived from an integrative synthesis of prior literature that shape EGCC. See Figure 1 for the conceptual framework.

Bandura (2018: p. 131) defines agency as “the ability to produce certain effects by one’s actions intentionally.” Thus, humans are not just reactive to environmental stimuli but can shape and regulate their current situation to achieve desired futures. Affordance theory is compatible with the notion of agency, as users intentionally actualize affordances rather than the technology simply possessing technical properties (Medeiros et al., 2025). Within the EGCC framework, agency enables employees to perceive GenAI’s affordances and to actualize them to support the development of creative outcomes through collaboration (He et al., 2023; Leonardi, 2011). SME employee agency represents a core mechanism that shapes how SME employees recognize, engage with and enact GenAI affordances in the creative process. Having agency, employees are likely to exercise reflection and awareness (Bandura, 2018), to facilitate meaningful experimentation and reflection on ideas with GenAI. These employees may be positioned to understand and utilize GenAI affordances for ideation and synthesis (Leonardi, 2011), critically evaluate and scrutinize its outputs, and actualize creative output through experimentation and adaptation (Frich et al., 2024; Rajaram and Tinguely, 2024).

In SMEs, employees’ agency is shaped by several factors. Multiple roles and resource constraints may prompt prioritization and personal initiative (Shin et al., 2012; Dwivedi and Pawsey, 2023) to actualize GenAI’s creative potential, or instead impose demands that limit active experimentation. Informal structures may increase autonomy and perceived agency, yet owner-managers attitudes towards acceptable GenAI practices (either positive or negative) may also enable or reduce agency in SME employees (Caldeira and Ward, 2002). As such, SME employees may have greater agency variance than employees in larger organizations with clear role boundaries and structural support. Lower SME employee agency may reduce their ability to recognize and act on affordances, resulting in standardized output and reduced innovation (Krakowski, 2025). Thus, SME employees’ agency operates as a micro-level driver within EGCC, shaping the extent to which they can effectively leverage GenAI to produce socially responsible creative outcomes.

We propose that SME employees can optimize the creativity benefits of GenAI when they possess a mix of competencies related to GenAI. GenAI competence refers to the technical, cognitive and reflective skills that help SME employees to perceive, interpret and realize the affordances of the technology. Based on affordance theory, competence here refers to the knowledge and skills that enable SME employees to translate affordances into creative outcomes. GenAI competence reflects SME employees’ ability to effectively recognize and actualize the action possibilities of GenAI in a purposeful, context-sensitive way. Zhao et al. (2025) show that higher-order skills, such as critical thinking, maximize the benefits of using GenAI. Similarly, Chowdhury et al. (2022, 2023) suggest that AI skills foster user trust and mitigate resistance as employees gain a better understanding of the technology, thereby sustaining adoption. Dwivedi and Pawsey (2023) note that competence also involves critically assessing and verifying AI-driven outputs, while applying expert judgment and cognitive engagement when interacting with AI systems. Madihalli et al. (2023) discusses AI-specific competencies for GenAI use, including prompt engineering, intelligent interrogation and fusion skills that balance human judgment with machine capabilities. These proficiencies can help SME employees complement and refine GenAI output, transforming generic output into creative, context-relevant work (Eapen et al., 2023). Thus, GenAI-competent SME employees will have the capabilities to effectively direct GenAI affordances and navigate the constraints of pursuing a partnership that can generate innovative outcomes.

Competency functions as a micro-level driver of EGCC, shaping how effectively SME employees can engage with and refine GenAI output to achieve novel, socially responsible creative outcomes. For example, an employee in a small retail store may possess strong prompt engineering skills to filter out generic or biased suggestions and adapt a tone that suits customer preferences. This competency enables the SME employee to harness the affordances of GenAI for EGCC by using the technology as a creative collaborator, rather than a text-production machine, thus helping the business achieve productivity gains. Without such competence, SME employees may fail to recognize or effectively leverage GenAI creative affordances, increasing the likelihood of superficial or standardized output, or output that may be misaligned ethically.

Domain knowledge refers to the depth of an employee’s knowledge of a particular field of study, market or professional context (McGuire et al., 2024). Domain knowledge will shape how SME employees interpret and actualize the affordances of GenAI. Within an affordance perspective, domain knowledge helps employees recognize which affordances are relevant and meaningful to enact in specific contexts, and that they also meet ethical and accountability standards. Employees with strong domain knowledge can provide contextual information to GenAI, accurately interpret the outputs, or critically evaluate the reliability and appropriateness of AI-generated content (Asamoah et al., 2024). GenAI tools achieve effectiveness by enabling users to integrate their personal expertise, such as product knowledge, industry standards and customer insights, into the decision-making process (Yang and Xu, 2025). Consequently, employees with extensive expertise are likely to be better able to work alongside GenAI systems, ensuring that the output aligns with organizational goals and ethical standards (Storey et al., 2025). Using domain knowledge to refine and contextualize GenAI outputs supports SMEs by ensuring innovation is relevant to local communities and ethically aligned. For example, a licensed electrician in a small construction SME can use GenAI to draft safety guidelines for client projects. The electrician’s technical electrical knowledge enables the employee to modify and contextualize the content generated by GenAI, producing a reliable and creative document. Without strong domain knowledge, the electrician may not be sufficiently critical of GenAI outputs, thus risking the use of generic or unsuitable content that may be superficial or violate ethical or professional standards, thereby constraining EGCC.

Task compatibility and reception refer to how well GenAI’s affordances match SME employees’ work tasks. Building on affordance theory (Gibson, 1977), we propose that creativity emerges when employees can recognize and apply tasks suitable for AI delegation that require human judgment and interpretation (He et al., 2023). He et al. (2023) find that users prefer to automate tasks with low process consequences (errors are expected and unimportant) and low social consequences (actions are autonomous and non-relational). In addition, they are more likely to entrust GenAI with familiar and moderately complex tasks. In the early or intermediate stages, such as for idea generation, production and design, GenAI’s affordances can help employees work faster and free them for interpretive and creative tasks (He et al., 2023) that depend on their professional knowledge and agency (Shamout et al., 2025). In contrast, GenAI affordances for front-stage work (He et al., 2023) and high-stakes tasks (Shamout et al., 2025), such as strategic insights or brand story building, may require employees’ professional knowledge and reasoning, as well as socio-emotional capacities. As such, Shamout et al. (2025) cautions that delegating tasks beyond GenAI’s capabilities can reduce trust and performance. Hence, we propose that SMEs calibrate GenAI involvement to the characteristics of the tasks and the stage of the work to enable SME employees to appropriately leverage GenAI affordances, such that GenAI can support ideation and generating options, while humans support interpretation, refinement and ethical evaluation. Within SMEs, task compatibility is crucial, as employees often have to manage multiple roles and tasks (Krakowski, 2025). This supports EGCC by allowing employees and GenAI to work on tasks where they excel, while remaining complementary (Medeiros et al., 2025). The alignment helps produce socially responsible creative outcomes by ensuring that SME employees use GenAI in contexts where outputs can be meaningfully used and refined in line with organizational and societal standards. This will help SMEs enhance their innovative capabilities while maintaining high-quality and ethical standards. For example, an SME employee at an interior design company may rely on GenAI to generate design images, then use their knowledge and expertise to finalize the layout, taking the client’s taste and local needs into account. Without compatible tasks, SME employees may over-rely on or underuse GenAI, limiting the realization of creative affordances for EGCC and compromising the quality and accountability of outputs.

Social-technical systems theory posits that employees rely on subsystems, such as social or technical systems, to operate and add value (Pasmore, 1988). We propose that the relationship between an organization’s social system (leadership, culture and ethics) and its technical system (digital infrastructure and AI tools) influences the creative collaboration of SME employees with GenAI. Socio-technical conditions shape the extent to which employees can effectively enact the potential actions of GenAI for experimentation and reflection, laying the foundation for EGCC. Leadership oriented towards AI demonstrates support for digital exploration and fosters psychological safety, enabling employees to engage confidently with GenAI tools (Zhao et al., 2025). Supportive climates and norms can also support how employees collaborate with AI technologies (Bankins et al., 2024). Specifically, people assess the use of algorithms performing certain job functions to determine whether they avoid or appreciate the technology. Technically, tangible AI capabilities such as digital infrastructure and data resources increase the organization’s creative capacity (Madihalli et al., 2023). In SMEs with limited resources (Krakowski, 2025), the presence of these socio-technical systems will be instrumental in transforming GenAI from a tool into a collaborative partner. Surprisingly, Kmieciak et al. (2012) provide empirical evidence that SMEs that foster a climate of empowerment do not find a relationship with innovativeness and likely require supportive and technological infrastructure, in addition. Overall, these supportive environments serve as foundational enablers that support or constrain SME employees’ enactment of creative affordances, shaping whether EGCC leads to novel, ethical and socially responsible creative outputs. Importantly, socio-technical and institutional conditions also directly influence social impact, rather than via EGCC. This is because the structural context can shape the extent to which stakeholders have access to and distribute digital resources, enabling inclusive access and equitable participation.

To illustrate EGCC in practice, we present three scenarios that show how SME employees enact creative affordances under different conditions of SME employee agency, GenAI competence, domain knowledge, task compatibility and reception, and socio-technical and institutional conditions.

A local ice creamery’s research insights team utilizes GenAI to design an online survey to investigate local ice cream flavours. The analyst identifies the affordances of GenAI for rapid ideation and drafting, then uses GenAI to produce initial survey items. The lead analyst, with knowledge and agency, revised the survey items to incorporate local slang and comply with local regulations, drawing on her understanding of beverage categories and local requirements. The SME has a culture of experimentation that fosters socio-technical conditions supporting the use of AI. The scenario illustrates how SME employees actualize GenAI’s creative affordances through a balanced act of collaboration. GenAI accelerates routine design work while the SME employee refines, contextualizes and checks the output. This balanced scenario illustrates the enactment of creative affordances, with GenAI supporting idea generation and efficiency, while human expertise ensures relevance and ethical compliance. The combination of SME employee agency, GenAI competence, domain knowledge and supportive socio-technical conditions leads to socially responsible creative outcomes.

A customer service officer at a cost-conscious skincare SME utilizes GenAI to create personalized communications for customers and product information. The SME employee is aware of GenAI’s capability to automate routine tasks but lacks the domain knowledge to accurately assess the factual accuracy of the results. The GenAI-generated product information was misleading and did not comply with advertising standards; as a result, the final output contained unverified claims. Despite a strong AI culture, the lack of oversight led to an overreliance on the tool. The SME employee only partially leverages creative affordances, as GenAI automation dominance has driven speed at the expense of authenticity and reliability. This imbalance illustrates how limited human agency and domain expertise can yield efficient but untrustworthy creative outcomes, thereby hindering effective EGCC. This scenario demonstrates an overreliance on GenAI with limited domain knowledge, which constrains the effective enactment of creative affordances for EGCC. While the SME employee achieves efficiency, the lack of human input leads to outputs that are not socially responsible, highlighting the risks of GenAI dominance.

A culinary marketer at a small native herbs-and-spices SME conceptualizes and designs new packaging and promotional images for a seaweed spice mix. The SME employee possesses high domain expertise and the agency to generate ideas through hand sketching and adjusting colour schemes based on his culinary and customer knowledge. The employee used GenAI to transcribe recipe notes and generate image drafts; however, this process does not fully utilize the technology’s visual development and rapid iteration affordances due to low GenAI skills and organizational support. The resulting output is authentic, but slow to produce. This results in the underutilization of the speed and diversity GenAI offers. Here, human expertise dominates, yet GenAI affordances are underutilized, resulting in lower creativity. The unbalanced partnership necessitates GenAI to serve as a co-creator of visuals, enabling faster image-based creativity and achieving sub-optimal EGCC. This scenario demonstrates the underutilization of creative affordances, where low GenAI competence and weak organizational support limited the ability to maximize GenAI for exploration and efficiency for EGCC. While the output was authentic and ethically aligned, the limited human-AI collaboration constrains the development of scalable creative outcomes.

This conceptual paper aims to address two research questions: RQ1. How do SME employees enact the creative affordances of Generative AI through Employee–GenAI Collaborative Creativity (EGCC)? and RQ2. What individual, task or socio-technical factors shape or constrain EGCC in producing socially responsible creative outcomes as a pathway to social impact? The paper proposes a theoretical framework for the key drivers of Employee-Generative AI Collaborative Creativity (EGCC) to help SMEs increase EGCC and achieve socially responsible human-GenAI creativity. The framework explains the actualization of GenAI affordances through the joint contributions of SME employees and GenAI to the creative process. This paper conceptualizes creativity as a dynamic, co-creative process in which SME employees and GenAI collaborate to generate, refine and evaluate ideas, aiming to achieve socially responsible creative outcomes. These outcomes serve as a mechanism for achieving social impact.

The framework positions EGCC as an outcome influenced by individual-level (employee agency and GenAI competence), task-level (domain knowledge and task compatibility) and system-level (socio-technical and institutional) factors. Answering RQ2, the proposed framework demonstrates how these factors at various levels enable or constrain EGCC in creating socially responsible creative outcomes, and that responsible innovation in SMEs requires a balance of human–GenAI collaboration, ethical guidance and the ability to adapt to context. Our theorization extends affordance theory to view GenAI as an affordance-enabling process for creativity, whilst emphasizing the role of SME employees in shaping the outcomes. Furthermore, we use a multilevel lens to show how affordances are determined. The fit among the individual, the task and the system can determine the actualization of GenAI’s affordances. This perspective highlights that SME employees actualize the creative affordances of GenAI through its interplay with human agency and competence, governance and organizational legitimacy, all of which support equitable work and inclusion, as well as accountability in the digital economy.

Our theoretical framework of key drivers focuses on SME Employee Agency, GenAI Competence, Domain Knowledge, Task Compatibility and Reception, and Socio-Technical and Institutional Conditions. We group them into three levels that emphasize affordance actualization as a multi-level phenomenon influenced by individual, task and system factors. More specifically, our model demonstrates that, at the personal level, agency and GenAI competence enable SME employees to perceive and act on GenAI affordances, thereby increasing their capacity to engage in meaningful work and participate equitably. At the task level, domain knowledge and task compatibility represent the degree to which GenAI affordances align with domain knowledge and task requirements, thereby helping to preserve authenticity and ethical standards that will maintain fairness and equity for customers and the broader community (e.g. industry). At the macro-system level, socio-technical and institutional conditions shape the legitimacy and responsible innovation through human-GenAI collaboration. This will directly or indirectly influence the social impact of well-being and access to digital technologies.

Building on the practical scenarios and the EGCC framework, SME managers can implement several strategies to integrate GenAI into creative outcomes. We structure these strategies into micro, meso and macro strategies. At the micro level, SME managers can support SME employees in developing human agency and GenAI competence by encouraging experimentation and training in skills such as prompt engineering, output interpretation and evaluation. Beyond training, managers can also facilitate checkpoints within creative tasks to allocate employees’ roles to validate output against organizational values or regulatory standards. SMEs can also foster peer learning to help employees exchange strategies and identify mistakes, thereby building collective knowledge to act on affordances. These practices help develop employee capabilities that translate into inclusive participation and meaningful work, which contributes to social impact at the micro level.

At the meso level, SMEs can leverage GenAI efficiency affordances by designing workflows that maximize automation with human refinement, such as GenAI expediting ideation and content generation. More structured human-GenAI work processes can be designed to assign employees responsibility for contextualization or regulatory compliance. In addition, SMEs can design workflows that align GenAI with specific task phases, such as using GenAI for early-stage idea generation and iteration. Specific phases involving employee input with strong domain expertise are important to ensure outputs are accurate and that they meet regulatory and industry standards. In doing so, SMEs co-create value through human-AI collaboration while ensuring creative outputs are socially responsible. Implementing these outputs can reduce harm (e.g. misinformation) and enhance access to resources or services (e.g. multilingual content). This can contribute to social impact outcomes for customers and the wider community, such as more equitable participation in services and improved well-being.

At the macro level, SME managers can create a supportive socio-technical environment that normalizes GenAI as a co-creator by actively integrating the technology to foster an experimental culture. Managers can communicate expectations for the use of GenAI transparently. Alongside this, managers can establish governance protocols to prevent misinformation and overreliance on the technology, such as conducting routine checks for accuracy, bias and alignment with organizational values. A participatory approach to everyday creative tasks can help normalize the role of GenAI as a co-creator. This may involve engaging SME employees to co-design workflows and facilitating team-based sessions to share insights and raise concerns. These interventions highlight the importance of socio-technical environments in shaping how EGCC translates to social impact for employees by reducing structural barriers to digital participation and ensuring inclusiveness.

Overall, these practices can enable SMEs to actualize GenAI’s creative affordances in an ethical and responsible manner, supporting Decent Work (SDG 8) and Innovation and Infrastructure (SDG 9). Building on the EGCC framework, we provide a practical tip sheet for SME managers seeking to translate affordance actualization into actionable strategies for socially responsible creative outcomes to achieve social impact. The tipsheet is presented in  Appendix.

Although the EGCC conceptual framework offers a fresh multilevel perspective for SMEs, it has limitations, and we call for further empirical work. We derive the framework from an integrative synthesis of prior literature on human–AI collaboration, creativity and socio-technical systems. Future research is needed to test the identified drivers of EGCC to attain creative and socially responsible outcomes, and to assess the extent to which EGCC generates measurable social impact. In number list below, we present future research questions derived from the EGCC framework, organized according to the five drivers that shape EGCC in SMEs.

Future research questions derived from the EGCC framework

1. SME Employee Agency:

   • How does SME employee agency shape the enactment of creative affordances in EGCC and with what implications for social impact outcomes?

   • How does SME employee agency influence the extent to which EGCC enhances employee well-being and meaningful participation in innovation processes?

   • In what ways does SME employee agency shape the balance between reliance on GenAI and human judgment in EGCC, and how does this affect the quality and inclusivity of outcomes?

2. SME Employee GenAI Competence:

   • To what extent does GenAI competence enable SME employees to move beyond surface-level use of AI towards engagement with creative affordances?

   • How does GenAI competence influence the fairness and quality of outcomes generated through EGCC that contribute to social impact?

   • To what extent does GenAI competence enhance the ability of EGCC to improve equitable access to services and digital technologies?

3. Specialized Domain Knowledge:

   • How do SME employees draw on domain knowledge to validate, adapt or reject GenAI-generated content within EGCC, and what implications are there for social impact?

   • How does the interaction between domain knowledge and GenAI competence shape the effective enactment of creative affordances in EGCC?

   • How does domain knowledge influence the access and relevance of outcomes generated through EGCC for diverse SME stakeholders?

4. Task Compatibility and Reception:

   • How do task characteristics (e.g. complexity, customer-facing vs back-end tasks) shape the enactment of creative affordances in EGCC?

   • How does task design shape stakeholders’ experiences of fairness, transparency and inclusion in outcomes generated through EGCC?

5. Socio-technical and Institutional Conditions:

   • How can SMEs design and manage EGCC to generate social impact while maintaining efficiency and competitiveness?

   • What organizational conditions (e.g. leadership, digital infrastructure) enable EGCC to improve access, inclusion and well-being in SME contexts?

   • What critical risks (e.g. bias, over-reliance on AI) can arise from EGCC, and how can SMEs mitigate these risks institutionally?

This  appendix offers a practical tool for SME managers to assess and improve the responsible use of GenAI in creative work. The tip sheet translates the SME Employee–Generative AI Collaborative Creativity (EGCC) framework into practical strategies for SMEs at individual, task and system levels. Managers can periodically review and update the checklist to ensure that SME employee-GenAI collaboration leads to impact for SME stakeholders and the overall system.