Surfing artificial intelligence: creating value in mission-driven contexts

In recent years, technologies such as artificial intelligence (AI) tools have evolved from an enabler of business to a critical driver of the sustainable and inclusive growth of organizations. By integrating technologies, organizations develop a business model that supports the creation and appropriation of value (Teece, 2010; Teece and Linden, 2017). More specifically, the adoption of technologies opens new avenues for capturing value, for example, gaining deeper insights into target needs, optimizing organizational processes and innovating the marketing offerings (Loureiro et al., 2021; Mancuso et al., 2023; Tavoletti et al., 2022). This innovation process helps organizations to respond quickly to market changes, offering new opportunities for value creation (Corsaro and Anzivino, 2021; Cucari et al., 2022; Sestino et al., 2024a, b).

Particularly, AI creates new opportunities for organizations to deliver added value to stakeholders by taking a proactive stance, effectively managing uncertainty, enhancing cost efficiency (Åström et al., 2022) and helping organizations to make more informed and strategic decisions (Peltier et al., 2024; Steiber and Alvarez, 2024). In this regard, recent data from a global survey highlight that in 2022, global corporate investment in AI reached approximately 92bn US dollars (Statista, 2024). In areas such as marketing, AI allows for the personalization of target experiences and the prediction of market trends, also facilitating resource management, optimizing processes and enhancing communication with target communities (AlNuaimi et al., 2022).

AI is revolutionizing not only the profit sector but also the public and nonprofit sectors, as it empowers organizations to be more agile, data-driven and target-focused (Sedkaoui and Benaichouba, 2024). Specifically in the third sector, technological evolution represents a profound change that enables nonprofit organizations (NPOs) to improve their efficiency (Gooyabadi et al., 2023), reach new beneficiaries and amplify the impact of their activities (Godefroid et al., 2024). More and more NPOs are implementing technologies to develop more efficient services and communication and establish relationships with stakeholders such as donors and beneficiaries (Gray et al., 2021; Ihm and Lee, 2021), as well as to improve the use of online platforms and social media (Gray et al., 2021). However, this innovation process also faces many challenges and tensions that need to be addressed (Jarvenpaa and Selander, 2023). This transition represents a complex and challenging path for NPOs (Masiero and Arvidsson, 2021; AlNuaimi et al., 2022).

Although the relevance of the AI implementation in organizations is widely recognized in innovation management studies, in the academic debate, there are few studies on the AI adoption by NPOs for marketing purposes (Brink et al., 2020; Vogelsang et al., 2021; Cipriano and Za, 2023). Indeed, as argued by scholars (e.g. Benbunan-Fich et al., 2020), in recent years, the topic of innovation management, understood as the adoption and implementation of new ideas, services or processes, has been little studied in NPOs (Do Adro and Leitão, 2020), while researchers have focused more on the private and public sectors. Moreover, as highlighted by the conceptual study by Sestino and De Mauro (2022), further studies are needed to clarify the role of AI in marketing strategies to infer trends, patterns and opportunities useful for improving the effectiveness of organizational processes. In an effort to fill this research gap, our study aims to investigate the role of AI in the marketing and innovation strategies of NPOs, with a particular focus on the value creation process within these organizations. We therefore propose the following two research questions:

Addressing these two research questions is also important in light of more recent data about AI adoption in the nonprofit sector. For example, data on the adoption of AI in British NPOs shows that they are at the early stages in fully understanding AI's potential, and that most of the organizations focus on generative AI because of its accessibility, its cost-efficiency and reduced complexity (Blacboud, 2024). The use of predictive AI is less common among British NPOs, as it requires greater resources and expertise. Similarly, in the Italian third sector, data on digital transitions highlight different levels of adoption of AI by NPOs (Italia Non-Profit, 2024): this overview reveals a wide spectrum of perceptions about digital transition, ranging from extreme resistance and slowness to more marked enthusiasm for this evolution. Moreover, more recent data from Techsoup (2025) show that 76% of NPOs interviewed don't have an AI strategy and 80% have no AI-acceptable use policy, but 85% have a high interest in tools like generative AI and predictive analysis and 47% of the organizations believe that AI could improve efficiency and productivity. However, 30% of the NPOs in the sample declared that financial constraints represent a barrier, and about 25% are worried about the social impact of AI.

The results of this empirical study contribute to the academic literature of innovation management by demonstrating how AI concurs to the value creation process of third sector organizations. In addition, the findings highlight the main challenges and critical issues which these organizations encounter in their process of adopting new technologies. Although scholars have focused on the role of technologies in capturing value (e.g. Bamel et al., 2024; Teece, 2010), to the authors' knowledge, this is one of the first studies which analyzes the value creation process in the third sector, considering the increasing and innovative role of AI.

The article is structured as follows. Section 2 is dedicated to the review of the literature on the topics of AI in the marketing strategies of organizations, and particularly the digital transformation of the third sector. In Section 3, the methodology adopted to answer our research questions is illustrated. In Section 4, the results are presented and discussed. Finally, in Section 5, conclusions, limitations and future developments of the research conclude the manuscript.

In recent years, the adoption of AI tools has attracted more and more attention from innovation management scholars, who have defined these technologies as pivotal drivers in the modern business landscape (Cannavale et al., 2025; Haefner et al., 2021; Mariani and Dwivedi, 2024). AI could be considered as a general-purpose technology based on new technical changes and with the potential to catalyze significant technological advancements, transforming skill requirements and reshaping the foundations of value creation (Brynjolfsson and McAfee, 2017; Barile et al., 2019). Value creation is considered a collaborative process that occurs in a specific context and is characterized by multiple relationships among heterogeneous actors (Vargo and Lusch, 2016).

In general terms, how to capture value from innovation is a widely discussed topic in innovation management literature. In this regard, the seminal studies of Teece (2010) have highlighted that technological innovation does not automatically guarantee the success of organizations, but it is necessary to adopt a contingent approach concerning how to organize the value chain. An effective design and implementation of the business model supported by strategic analysis is necessary for technological innovation to be successful. Organizations operate in a complex innovation context, and capturing value by implementing innovative business models is increasingly difficult (Teece and Linden, 2017). In this sense, AI tools fit into the value chain by supporting organizations. For example, AI could add value in processing and interpreting large datasets to support human decision-making, generating ideas and solutions and analyzing potential outcomes (Huang and Rust, 2022). AI is considered a tool which analyzes data, monitors processes in real time and executes programmed tasks, also able to interpret emotions (Huang and Rust, 2021). AI tools offer organizations many significant advantages and opportunities to ameliorate process automation, market forecasting and decision-making processes (Volkmar et al., 2022). AI also generates value through the provision of personalized real-time recommendations, service improvement and personalized response to customer needs (Davenport et al., 2020).

Although the benefits of implementing AI have been widely recognized in academic debate, the adoption of these tools within organizations is not a linear and immediate process but presents obstacles and challenges for organizations (Akter et al., 2023; Füller et al., 2022). In particular, innovation management scholars have also focused on the dark side of AI, highlighting a series of uncertainties about the possible implications and consequences of these technologies (Belanche et al., 2020). For example, according to Davenport and Ronanki (2018), AI is too risky or expensive to use when organizations are less competent and do not fully understand how to manage it. Indeed, the process of implementing AI tools requires specific innovation capabilities, that is, capacities which organizations possess and cultivate to innovate (Camison and Villar-Lopez, 2014). AI tools influence these innovation capabilities of organizations (Gama and Magistretti, 2025), leading to the enhancement or automation of decision-making (Raisch and Krakowski, 2021) and product development (Kellogg et al., 2020).

Similarly, the widespread resistance among employees and customers to the use of AI is crucial for organizations. In this vein, the findings of Rasheed et al.’s study (2023) have revealed that customer values and global motivations play a key role in the adoption of AI tools. Factors that promote AI adoption include perceived ease of use, perceived usefulness, perceived pleasure and perceived innovation; conversely, factors that inhibit the adoption of AI tools include technological complexity, technological anxiety, perceived security and privacy concerns. Indeed, resistance to adopting AI is often linked to ethical dilemmas and questions of accountability (Tóth et al., 2022).

As for these latter ethical aspects related to the implementation of AI, recent studies have shown that organizations are increasingly aware of the ethical issues that these technologies can cause, but at the same time, they are willing to engage and take responsibility for intervening (Macnish et al., 2019). In this context, organizations are taking measures to mitigate the ethical issues of AI, although these often go beyond their competencies (Stahl et al., 2022). For example, these challenges can be mitigated through enhanced human oversight, which in turn fosters greater acceptance (Horvath et al., 2023). In addition, certain system-level characteristics, such as high accuracy, accessible appeal processes, transparency, affordability and data privacy (Volkmar et al., 2022), can help reduce resistance (Horvath et al., 2023). As argued by Leung et al. (2018), an extensive and improper use of AI tools and big data by senior managers could generate tensions with the other managers, as middle managers, who might perceive themselves as undervalued and misunderstood. Furthermore, resistance may arise when automation constrains the capacity of customers to demonstrate competencies, potentially alienating key customer groups (Leung et al., 2018).

Today, the rapid growth, spread and social acceptance of AI technology is developing a paradigm shift that extends beyond simple technological innovations (Stahl and Eke, 2024). As technology becomes more than a tool, but also a learning instrument (Zheng et al., 2017), value is created by the interaction between humans and machines, and not only in human-to-human interactions (Barile et al., 2024). In the context of human–machine interaction, some scholars argue that implementing and utilizing digital tools demands human skills and expertise to interpret data and make it actionable for management purposes. Therefore, it is essential to adopt a holistic perspective that integrates both technological and human factors (Corsaro and D'Amico, 2022).

In particular, human–nonhuman interactions play a crucial role in the adoption of AI within NPOs (Holzer, 2024). For these organizations, as recently argued by Plaisance (2025), AI presents both significant opportunities and serious risks. For example, Fine and Kanter (2020) have highlighted two realistic benefits, i.e. freeing up time for human-focused work and fostering collaboration through shared tools. However, challenges include concerns about AI promoting for-profit-like models, the difficulty NPOs face in generating quality data, the risk of dependency on commercial providers and the need for responsible use. In this vein, some researchers examined the advantages of collaboration and hybrid intelligence and demonstrated the potential of combining the strengths of human and AI intelligence (Bouschery et al., 2023; Sedkaoui and Benaichouba, 2024). In this research stream, some studies underlined the importance of interactions and integration between humans and AI in terms of redesigning human operations to align with AI's capabilities and to reshape outcomes and dynamics (Mikalef and Gupta, 2021; Saenz et al., 2020; Simòn et al., 2024).

In particular, Simòn et al. (2024) argued that value derived from AI increases when a fruitful dialogue improves interoperability by promoting complementarities, recognizing the importance of data and the definition of data requirements, and considering also organizational constraints. Another key issue relates to trust: AI's value increases when dialogue builds trust, focusing on the tangible advantages of technology and transparency. In addition, a constant and productive dialogue between humans and AI promotes mutual understanding and knowledge through resource specialization and a data-based culture. All these considerations open the path to understanding and focusing on the risk propensity and are open to change when dealing with technological innovations (Sestino et al., 2024a, b). Moreover, the degree of risk-taking linked to the uncertainty typical when a new technology is adopted (Rogers et al., 2014) and its impact on the nonprofit sector takes on different nuances when considering such a specific field as the one being studied here.

Recent studies revealed that there are numerous opportunities to use AI technologies in NPOs' marketing strategies (e.g. Jaskyte et al., 2018). AI could support marketing managers in decision-making activities, helping them in analysing huge amounts of data (Du and Xie, 2021). Human–AI collaborations go beyond a simple and static interaction between a person and a tool: the effectiveness of an AI-powered Customer Relationship Management (CRM) system, for example, depends on the manager's capacity and willingness to input specific data and train the system (Gaczek et al., 2023) and on the willingness of individuals' to take risks (Sestino et al., 2024a, b). Specifically, people more risk-inclined consider the potentialities in terms of improvement offered by new technologies (Meertens and Lion, 2008), and persons more open to change consider the innovation and the transformative changes of these technologies.

In this innovative and integrative process between human and AI technology, there are also some common hesitations in trusting AI and its tools – people usually rely on algorithmic advice when considering objective or numerical issues (Castelo et al., 2019). Some managers are cautious in fully adopting AI for decision-making despite their potential (Davenport and Kirby, 2016), also because AI tools do not explain what they have done and how they make their decisions (Kolbjørnsrud et al., 2016). Managers need to reconsider interactions between humans and technology to fully benefit from AI, also considering strategy, ethics and psychology (Volkmar et al., 2022). This process requires a redesign of roles and responsibilities between humans and machines (Hoffman and Novak, 2018).

Digital transition, and in particular AI, holds significant potential to enhance the efficiency, transparency and accountability of organizations. However, implementing digital tools within NPOs, characterized by a social and humanitarian mission and objectives, must align with ethical principles, which necessitate a structured approach with well-defined design principles and guidelines to ensure ethical use (Baharmand et al., 2021). Given the financial constraints these organizations often face, NPOs prioritize operational efficiency to maximize the impact of limited resources (Comes et al., 2018).

However, the strategic development of digital transformation within NPOs and its impact on marketing and fundraising is still not well understood (Vogelsang et al., 2021), due to limited knowledge about the prerequisites and value creation frameworks required for digital transition in these organizations (Brink et al., 2020; Cipriano and Za, 2023).

In conclusion, the main findings emerging from the literature on the role of AI in value creation, and specifically in NPOs, are summarized in Table 1.

As demonstrated in Table 1, there is a plethora of literature that contains a multitude of diverse and fragmented debates. The potential of AI to reshape markets and act as a pivot in the value creation process is well-documented (see Barile et al., 2019; Teece and Linden, 2017). However, it is also fundamental to consider organizational capabilities and business models. Furthermore, when contemplating AI, it is imperative to deliberate on the opportunities and challenges inherent to the field (see Huang and Rust, 2021; Atker et al., 2023). Another significant and ongoing discourse pertains to the role of human–AI collaborations (see, for example, Bouschery et al., 2023; Simón et al., 2024). These debates, although still in their infancy, are also vibrant in the nonprofit sector: As previously stated, AI offers numerous opportunities but also many risks. The strategic value of digital transformation for innovation and marketing is still underexplored (see Cipriano and Za, 2023).

Given this, it is clear that there are some research gaps: little is known about how AI adoption in NPOs simultaneously shapes value creation, stakeholders' relations and engagement and ethical and privacy issues. In addition, little is also known about NPOs in the management of human–AI collaborations. The objective of this study is to advance, through qualitative research, academic research on these topics and to provide practical guidance to managers on the management of digital transformation in the nonprofit sector.

The objective of this empirical research is to analyze the adoption of AI in the strategies of NPOs, with a focus on its role in the value creation process. Given this exploratory research objective, we opted for a qualitative approach based on the collection and analysis of semi-structured interviews with third sector professionals with experience in AI adoption. Semi-structured interviews were chosen to allow flexibility in exploring participants' perspectives while ensuring consistency between the core themes of this study. The method of semi-structured interviews is, in fact, widely adopted in academic literature to analyze the phenomenon of digital transformation (e.g. Visvizi et al., 2022) and, in particular, to investigate the role of AI in organizations (e.g. Steiber and Alvarez, 2024).

Specifically, for our research, we employed the “systematic combining” method, rooted in an abductive approach, which is particularly effective for developing new theories by integrating a theoretical framework, empirical fieldwork and case analysis (Dubois and Gadde, 2002). In this method, the presence of an analytical framework is crucial. As Miles (1994) advised, the framework should not be overly rigid or prestructured, as this could distort the perspectives of informants. At the same time, it should not be too flexible to avoid indiscriminate data collection and information overload. The abductive approach facilitates a seamless interplay between theory and practice, a process we have adhered to throughout.

As for what concerns the selection of sources, it was based on theoretical relevance and, as the approach requires, continuously refined to answer the emergence of new themes and constructs during the research process.

The research setting for this study focuses on NPOs, an underexplored domain in academic literature on digital technologies adoption in marketing strategies (Godefroid et al., 2024). The third sector today is characterized by a strong demand useful for enhancing fundraising, engaging donors and streamlining operations. This sector is a useful setting for our research, as data from a survey carried out in June 2023 in the USA showed that more than 50% of marketers who are using AI tools in their marketing programs recognize the leading benefit of the increased speed or efficiency of workflows or processes and their benefits in data processing (Statista, 2024).

Specifically, our research setting offers a valuable perspective on how AI tools can be leveraged by organizations, characterized by a prominent social mission, in their innovation process, and could improve their digital transformation. In this paper, we focus on the Italian context as the Italian nonprofit sector, usually characterized by slow innovation, is exploring AI to improve relationships with users and donors, and to optimize resource management. Research conducted by Italia Non Profit (2024) has underlined that more than 40% of Italian NPOs are on the path of digitalization, but without a strategic approach, and about 13% have stated that they have fully incorporated digital tools into their organization's operations and culture.

This sector is experimenting with innovative solutions to facilitate interactions and to make responses to users' needs more efficient (Bva Doxa, 2024). Specifically, the nonprofit sector in Italy is beginning to grasp the potential of these new digital tools in personalization and communication, even if the digitalization of this sector shows an overview with different levels of adoption.

The data collection process was conducted through 16 semi-structured interviews with 13 professionals from 9 NPOs, a research center, a communication agency and a freelance consultant with experience in managing digital projects of NPOs (Table 2).

Adopting the theoretical sampling approach, the 13 key informants were selected for their expertise in the field, providing a sample that was both knowledgeable and diverse. Some informants were interviewed multiple times. According to Ligita et al. (2020), theoretical sampling serves to clarify and illuminate the variations, properties, dimensions and relationships between codes and categories to develop a credible and authentic theory. A common practice among grounded theory researchers to assess the effectiveness of categories derived from the data is to apply theoretical sampling. As Charmaz (2014) states, the process of conducting theoretical sampling relies on the prior identification of a category, allowing researchers to refine their theoretical categories further. This selection method enabled the gathering of rich qualitative data, capturing subtle insights and supporting the identification of patterns and shared experiences among participants.

We selected key informants able to communicate and discuss the object of our study (Kumar et al., 1993) with different backgrounds to argue and examine similarities and differences in the phenomenon of interest (Patton, 2002). We adopted the following selection criteria to ensure the contributions of key informants' to our study (Saunders and Townsend, 2018; Miles et al., 2013):

1. professional seniority and experience in the third sector: key informants were required to hold senior positions, such as senior managers or consultants, and possess long experience in the third sector, ensuring they had a comprehensive understanding of its dynamics;

2. relevant project leadership: key informants should have direct experience managing or coordinating projects related to marketing, communication and/or fundraising. This criterion ensured that they had engaged with relevant strategic and operational decisions;

3. use of AI tools: key informants had to have practical experience using AI technologies within their organizational roles. This ensured they could provide grounded reflections on how such tools are applied in practice, their benefits and limitations, and their implications for the sector.

We have excluded potential key informants too young and/or with no sufficient experience or understanding of the nonprofit context. Moreover, we have not included managers not involved in the management and coordination of marketing, communication or fundraising projects, and those not involved directly or indirectly in innovation processes.

Semi-structured interviews were conducted by both researchers in the period between May and November 2024 via video calls. The interviews were recorded and transcribed verbatim, and lasted 60 min on average; the researchers also took notes during the interviews.

Specifically, each interview was guided by a set of predefined questions but allowed for open-ended responses, enabling key informants to discuss their experiences, challenges and viewpoints in detail. Indeed, an interview protocol (see Table A1 in Appendix) was developed by researchers based on the three following macro-themes retrieved from the analyzed academic literature:

1. The role of AI in achieving marketing and communications objectives in organizations;

2. The implementation of AI for fundraising purposes;

3. The adoption of AI to develop relationships with stakeholders and increase their participation.

This interview protocol was adopted for the interviews with the key informants from NPOs; for the key informants from other organizations (e.g. multinational communications agency, research center, etc.), we adapted the questions to investigate how the third sector is implementing AI tools from their perspective.

During the interviews, some influencing variables, for example, experience in digital transformation, seniority and organizational role, has been considered to ensure methodological rigor and accurate interpretations of the findings. The data collection phase ended when the saturation point was reached, that is, when the themes recurred in the transcriptions of the previous interviews and no original elements emerged from the last interviews (Aldiabat and Le Navenec, 2018; Mwita, 2022). Specifically, as we conducted interviews based on the saturation principle suggested by Guest et al. (2006), we interviewed key informants until no new insights emerged: after 11 interviews, data saturation was reached, and by the 14th interview, the most important themes had been acknowledged. The use of the saturation principle ensured the effectiveness of the study and the avoidance of replications, ensuring a holistic viewpoint of the object of the study (Thomas, 2024).

During the data collection, anonymity was guaranteed to reduce social pressure and to protect identities (Coffelt, 2017).

In addition, primary data were triangulated with secondary data: secondary data included press articles, archival documents, reports and other documents shared by key informants involved in this study. The authors analyzed the various data independently to ensure triangulation and mitigation of bias (Yin, 2013) and to ensure the validity of this first phase of data collection. Furthermore, the adoption of a protocol assessing the scheduling, the recordings and the follow-ups of the interviews ensured the reliability of the process.

Key informants' potential bias has been guaranteed by granting the anonymity to informants to encourage them to discuss openly during our interviews (Eisenhardt, 1989a, b), providing open-ended questions (Koriat et al., 2000) and triangulating data from interviews with a wide range of secondary data (Bingham and Eisenhardt, 2011).

The interview transcripts represent the main, but not the only, material used by the researchers in the data analysis phase of this study. Particularly, the coding of semi-structured interview transcripts was conducted using Gioia et al.'s (2013) methodology, which is a systematic and rigorous approach widely adopted in qualitative research for its ability to develop grounded theories directly from data.

This multistage process began with an open-coding phase, during which each interview transcript was meticulously reviewed to capture first-order concepts. At this stage, particular attention was paid to the use of the language, expressions and specific terms of the participants. This initial phase is intentionally descriptive, aiming to gather a comprehensive set of insights without imposing any predefined theoretical structures, thus allowing the participants' perspectives to lead the analysis. A key aspect of this phase involved reviewing each segment of the transcripts repeatedly to ensure that even nuanced details and unique viewpoints were represented among the first-order codes. During this phase, the researchers conducted the analysis independently and discussed misalignments in interpretations: after reviewing the multiple codes several times, they grouped overlapping and similar themes and discarded codes that were not relevant to the research aims. Finally, they came up with 20 first-order concepts.

Following the identification of a broad set of first-order concepts, the analysis moved to the second stage, focused on grouping these concepts into second-order themes. This phase required a continuous process of reflection, comparison and refinement. By examining the relationships between first-order concepts, broader and more abstract themes were distilled, which often revealed patterns and recurring ideas that extended beyond individual participant responses. These second-order themes emerged as interpretive categories, representing shared experiences, challenges or insights across participants. During this stage, iterative coding was crucial, with the researcher frequently reviewing the transcripts to ensure that the themes accurately captured the breadth and depth of the data. In addition, in this phase, the researchers conducted the analysis independently and, after discussing heterogeneous and sometimes not aligned interpretations, nine second-order themes were defined. Practically, the first-order concepts (e.g. “AI helps in crafting project proposals to obtain funding” and “AI identifies new actors who could support the social cause of the NPO”) have been grouped in a more abstract theme (e.g. “Enhancement of new funding opportunities”) based also on literature.

The final stage of the Gioia method involved synthesizing the second-order themes into aggregate dimensions, which represent the core conceptual insights from the study. These dimensions serve as the highest level of abstraction, consolidating the data into a coherent framework (Figure 1). Specifically, second-order themes have been grouped into three aggregate dimensions: AI for marketing communication objectives, AI for fundraising purposes and AI for stakeholder engagement.

Throughout the data analysis phase, researchers regularly documented thoughts, interpretations and coding choices to enhance analytical clarity and rigor. Furthermore, cross-comparisons were made between transcripts to ensure consistency, with discrepancies analyzed in detail to refine the understanding of each theme. By following the Gioia methodology, this coding approach allowed for a nuanced, layered analysis that maintained fidelity to the participants' original viewpoints while yielding meaningful insights relevant to the study's aims. The result was a structured framework that reflects the complexities of the research topic.

This systematic approach not only ensured that the analysis was thorough and deeply informed by the key informants' experiences but also laid a robust foundation for the study's theoretical contributions, offering a comprehensive view of the underlying patterns, challenges and insights revealed through the data. The Gioia method thus provided a path for developing findings that are both richly descriptive and conceptually insightful, contributing valuable understanding to the field.

The reliability and validity of the study have been ensured using appropriate methodology and analytical data processing, ensuring logical consistency among data, analysis and findings (Carcary, 2009). Moreover, our key informants have been informed about the aim of this study, and we discussed with them the themes needing to be clarified and the researchers' interpretation to ensure their perspectives.

The results of our empirical analysis focus on how NPOs adopt AI technologies into their innovation strategies, focusing in particular on marketing and communication, and their role in the value creation process. Particularly, our research demonstrates that NPOs implement AI to achieve marketing communications objectives (Section 4.1), for fundraising purposes (Section 4.2) and, finally, to increase stakeholder engagement (Section 4.3).

The adoption of AI in achieving NPOs' marketing objectives is increasingly relevant. For example, according to a recent study conducted by Stanford University (2024), approximately 50% of nonprofit respondents report using AI technologies, mainly for supportive tasks such as operational processes in finance, human resources, contracting and marketing.

In this vein, our research highlights that AI is used by NPOs both strategically, for long-term planning aimed at achieving specific organizational goals, and operationally, for example, in practices to be carried out in the short or medium term, such as the daily management of the organization's digital channels. Thus, AI plays a strategic role in helping NPOs achieve their marketing communication goals by enhancing data-driven decision-making, personalizing outreach and improving engagement with their target audience. This dual role of AI was highlighted by the Digital Specialist interviewed (code R.3.1):

More specifically, AI enables NPOs to better understand donor behavior, predict trends and optimize communication campaigns. This is further supported by secondary data (Stanford University, 2024), indicating that a significant number of organizations are also leveraging AI for mission-driven initiatives. Moreover, around 75% of respondents believe their organizations could greatly benefit from expanded AI use, particularly in areas related to their core mission (Stanford University, 2024).

By leveraging AI tools, NPOs can efficiently allocate resources and create more impactful, targeted marketing communication strategies, ultimately increasing their social impact and reaching key organizational objectives. A key informant (code R.8) stated:

In this regard, the coordinator for nonprofit studies (code R.10) also confirmed the key role of AI at a strategic level, but also underlined the role of human–AI interactions as considering these tools as complements useful to guide the decision-making process:

Regarding the operational aspect, our study highlights that the organizations analyzed exploit the ability of AI mainly to automate repetitive and routine tasks, but at the same time require creativity and analytical skills. For example, our key informants argue that AI is used as an automatic writing tool to develop textual content such as article drafts, service or event descriptions, newsletters and social media posts. In addition, secondary data show that 23.7% of NPOs are beginning to use marketing automation tools to automate some marketing tasks, enhance engagement and manage campaigns (Techsoup, 2025).

Furthermore, AI is adopted for graphic purposes: it plays a key role in the design process and the creation of posters, images and videos for the various physical and digital touchpoints included in the marketing communication strategy of NPOs. Nevertheless, our key informants also specified the importance of guiding AI tools in designing communication campaigns and in redefining the requests after the first AI proposals. Specifically for social media management activities, AI is being increasingly used to improve efficiency and effectiveness. AI supports organizations to analyze large volumes of data to identify trends, monitor target engagement and tailor content to specific communities. AI-powered tools can automate tasks such as scheduling posts, responding to inquiries and generating content suggestions based on user preferences. By leveraging AI, NPOs optimize their strategies in real-time, ensuring consistent, personalized communication with followers, improving the interaction with stakeholders, and finally driving engagement and brand awareness.

However, although the potential of AI in marketing communication strategies is widely recognized by our key informants, the results highlight a series of barriers hindering its adoption. For example, according to the coordinator for nonprofit studies (code R.10), there is significant internal cultural resistance in the nonprofit sector toward the adoption of new technologies, particularly AI:

Generally, the lack of digital skills in NPOs poses a significant barrier to adopting AI technologies. Many NPOs struggle with limited resources and face challenges in recruiting or training staff with the technical expertise needed to leverage AI effectively. This skill gap hinders the ability of NPOs to integrate AI into their operations, preventing them from fully benefiting from data-driven insights, process automation and enhanced marketing communication strategies. Consequently, the potential impact of AI in increasing efficiency and outreach within the third sector remains largely untapped, considering that secondary data highlight the presence of limited internal AI expertise (Techsoup, 2025).

Fundraising is crucial for NPOs, as it provides the financial resources necessary to sustain their operations and achieve their mission. For example, effective fundraising allows NPOs to fund projects, support communities and expand their reach. It also helps to build relationships with donors, creating a network of supporters interested in the mission of the organization. Without adequate fundraising, NPOs may struggle to meet their goals or even remain operational. Therefore, developing strong fundraising strategies is essential for long-term success and impact in the third sector.

Nowadays, AI is transforming fundraising strategies of NPOs by providing advanced tools which enhance efficiency, personalization and engagement. Data from a global survey show that more than 50% of NPOs have not started planning to use AI for fundraising, but about 15% aim to adopt these tools within the next six months (Techsoup, 2025).

One of the primary ways AI supports fundraising is through data analysis – by processing large volumes of donor data, AI can identify trends and patterns in giving behavior, enabling NPOs to segment their donor base more effectively. AI also plays a key role in automating routine tasks, such as sending personalized e-mail campaigns, managing donation appeals and scheduling follow-up communications. This automation not only saves time and resources but also ensures that donor communications are consistent and timely, enhancing the overall donor experience. AI-powered chatbots can provide immediate support to donors, answering questions, guiding them through the donation process and providing real-time assistance. In addition, AI optimizes fundraising campaigns by predicting which donors are most likely to give and how much they are willing to contribute. It can also suggest the best times to reach out to donors, ensuring that messages are received when they are most likely to lead to a donation. With machine learning algorithms, AI can continuously refine and improve fundraising strategies based on the success or failure of past campaigns, leading to increasingly effective approaches over time. The use of effective prompts, which require some knowledge of tools and human–AI collaboration, could be useful to obtain detailed information on key aspects of the target donors. For example, one of the key informants suggested that the use of a clear, specific and above all detailed prompt allows AI to identify motivations and psychological barriers, providing emotional insights that enrich campaigns. In this sense, AI not only supports a better understanding of donor profiles but also plays a strategic role in influencing their behavior. By analyzing sentiment, past donation patterns and engagement history, AI can help tailor persuasive messages and suggest the optimal timing and channels for communication. This enables NPOs to craft more targeted and emotionally resonant appeals, while increasing the likelihood of donor action and strengthening long-term commitment.

Moreover, to secure funding, NPOs use AI not only in their relationship with donors but also to develop projects that are to be submitted to institutions for public and private funding opportunities. AI helps streamline the process by analyzing trends, identifying potential funding sources and assisting in the creation of well-targeted project proposals which align with the goals and requirements of funding bodies. In this regard, the Innovation Manager we have interviewed (code R.5) stated:

Hence, by leveraging AI, NPOs maximize the impact of their fundraising efforts, improve donor retention and, ultimately, raise more funds to support their missions. The ability to engage donors more personally and efficiently enables NPOs to build long-term relationships, ensuring sustainable financial support for their causes. However, our research findings highlight that NPOs often face significant barriers in adopting AI tools, largely due to a lack of internal expertise and technical skills. Many NPOs operate with limited budgets, prioritizing funds for immediate mission-driven activities, which often do not consider investing in advanced technologies like AI. This financial constraint is compounded by a shortage of staff with the necessary technical backgrounds to implement and manage AI solutions effectively. As a result, NPOs struggle to understand the potential benefits and risks of AI, limiting their ability to make informed decisions about its integration. Our key informants suggested that to bridge this gap, accessible training and educational programs tailored to nonprofit contexts are essential: these programs can equip NPO professionals with foundational AI knowledge and practical skills, empowering them to harness AI's potential for enhancing their impact while remaining mission-focused. The central role of training in the effective use of AI tools to support fundraising activities has also been recognised at the government level. For example, the Charity Commission for England and Wales (2024) has produced guidance for third sector organizations on how to raise money for the charity by reaching out to trustees.

From our research, it emerged that the implementation of AI in NPOs offers an opportunity to enhance stakeholder engagement by enabling more strategic and data-driven approaches. Specifically, our results highlight that AI plays a pivotal role in transforming engagement strategies within NPOs by enabling a more dynamic, personalized and proactive approach to interacting with stakeholders. By leveraging AI, NPOs analyze vast amounts of data to understand individual behaviors, preferences and engagement patterns in real time. This insight allows for the creation of highly customized content and communications that resonate deeply with each stakeholder, making every interaction feel relevant, timely and impactful. For example, AI tools suggest content, updates or volunteer opportunities based on each stakeholder's unique interests, not only boosting engagement but also strengthening their connection to the organization's mission. In this regard, the key informant R.4 stated that:

According to The Non Profit Times (2024), 90% of NPOs are utilizing AI across various engagement and marketing applications. For example, these include constituent engagement, contact center operations, survey platforms and customer analytics. Moreover, 68% of NPOs indicated using AI to analyze user data and gain insights into their needs and challenges, slightly surpassing the 64% of B2C brands doing the same. According to this report, 87% of NPOs emphasized the importance of robust digital engagement for effectively reaching, enrolling and serving new users. To achieve these goals, they are increasingly relying on AI to enhance and accelerate their efforts.

In addition, our key informants argued that AI-based tools, such as chatbots and virtual assistants, have redefined stakeholder interaction by providing instant support and information. These tools do not just answer questions; they enhance the user experience by reducing wait times, providing precise information and addressing multiple inquiries simultaneously. AI tools even recognize sentiment, enabling them to respond empathetically and adding a personal touch even in automated responses. This level of support keeps stakeholders engaged, encouraging deeper, ongoing involvement with the NPO.

Furthermore, by understanding what stakeholders are likely to need or want in the future, NPOs proactively reach out with relevant opportunities, resources or updates before a request is made. This proactive approach not only sustains engagement but also builds trust, as it demonstrates the organization's commitment to understanding and supporting its community.

In conclusion, our findings clearly show that AI allows the analyzed NPOs to adjust their engagement strategies in real-time, testing and optimizing campaigns quickly. However, as confirmed also by secondary data, larger NPOs use AI to enhance engagement, while smaller ones are exploring AI tools with a slower step (Techsoup, 2025). This adaptability makes it easier to maintain high engagement levels, as AI continuously refines content, timing and delivery based on user feedback and behavior. Thus, AI empowers NPOs to create more meaningful, engaging and lasting relationships with their stakeholders by transforming how they understand, reach and interact with them. By making each engagement more relevant, responsive and proactive, AI helps NPOs build loyalty, trust and sustained support through a seamless, personalized experience that aligns with their mission and goals.

The objective of this research is to investigate the phenomenon of AI adoption in the strategies of NPOs and, in particular, in their value creation process. The results of the study reveal that AI is implemented in marketing strategies mainly for three purposes: i.e. (1) to achieve marketing communications objectives, (2) to raise funds and (3) to increase stakeholder engagement. The findings allowed us to develop a framework in which the role of human–AI interactions in the three aforementioned macro-areas emerges at both a strategic and operational level (Figure 2).

The framework presented provides a comprehensive and holistic structure for considering human–AI interactions at strategic and operational levels to effectively achieve marketing communications goals, improve fundraising campaigns and enhance stakeholder engagement.

More specifically, at the strategic level, our results focus on long-term objectives that drive the success of marketing and fundraising initiatives. Adopting an external viewpoint, in line with the literature (Hoffman and Novak, 2018), our findings suggest the importance of considering human–nonhuman interactions, as some of the key informants from NPOs have underlined the redesign process they have activated in their organizations to redefine roles and tasks. Furthermore, at a strategic level, the alignment of marketing communication strategies developed with the support of AI requires constant interaction with the marketing professionals of NPOs. For example, our results highlight that human intervention guarantees an alignment of values between what is proposed by the AI and the mission and vision of the NPOs and their donors. From this perspective, our study confirms that the adoption of technologies in organizations does not automatically allow for to capture of value (Teece, 2010), but the process of adopting these tools requires a rethinking of organizational competencies. Indeed, as argued by Du and Xie (2021), AI supports marketing professionals during their decision-making process by complementing their human judgment with data. More specifically, at this strategic level, AI incorporates predictive analysis to anticipate trends in donor behavior and market conditions, as well as fostering relationship-building strategies aimed at cultivating trust and loyalty among stakeholders, which are fundamental to sustaining long-term engagement. AI tools enable advanced target segmentation (Choe et al., 2024), enabling organizations to identify and cater to specific audience groups with precision; resource optimization, ensuring that efforts and investments are efficiently allocated for maximum impact; and the creation of strategic campaigns that align with broader organizational goals.

Meanwhile, at the operational level, our findings detail the practical tools and processes necessary for executing these strategies and creating value for NPOs by supporting them in day-to-day activities. Central to this are technologies and methodologies for the automation of communications, which streamline the delivery of messages across various platforms, and content optimization, ensuring that the messaging resonates effectively with the target audience. Social media management through AI plays a crucial role here, leveraging digital platforms to enhance NPOs' visibility and interaction, while monitoring and performance analysis provide real-time insights to track the effectiveness of campaigns and make data-driven adjustments (Dwivedi et al., 2021). The operational focus also includes the ability to guide users through the donation process with tailored interactions, monitor stakeholder feedback to remain responsive to their needs, and analyze data to produce clear, actionable reporting (Matthews et al., 2024). Indeed, by enabling the verification of donors, organizations and beneficiaries, AI allows donations to reach intended recipients more swiftly, which is especially critical in emergencies. This streamlined process not only accelerates the delivery of aid but also enhances public confidence in donation systems by providing greater assurance that funds are going to the right people (Hu and Li, 2020). When donations reach beneficiaries, digital platforms automatically notify donors, creating an added layer of transparency and reinforcing trust in the process. Furthermore, funds cannot be accessed until they receive approval from authorized personnel, ensuring that resources are used appropriately and for their intended purposes (Sahebi et al., 2020). However, this platformization could also show many problems in this context related to value creation in a digital setting: as already mentioned in B2B literature (Corsaro and Anzivino, 2021), if the process is not properly managed, value could not be created, and this could affect the efficiency of the introduction of AI tools in NPOs.

The integration of these two levels, i.e. the strategic and the operational ones, allows NPOs to anticipate donation trends, create strategies that align with future demands, and continuously refine their approaches based on feedback and performance analytics (Park et al., 2024), thus playing a key role in the value creation process of NPOs. Particularly, our results highlight that AI tools are adopted by NPOs first at the operational level and then at the strategic level. Together, the strategic and operational levels ensure a balanced approach that not only sets long-term goals but also provides the tools and insights to adapt dynamically, driving stronger engagement, loyalty and successful outcomes. According to literature (Simòn et al., 2024), this balanced approach is sustained by a constant and complementary dialogue between the human and AI tools. This holistic perspective underscores the need to balance technological innovation with a deep understanding of human factors, ensuring that the AI tools employed are aligned with the organization's objectives and the stakeholders' expectations. In this vein, compared to previous studies on innovation management in the third sector that have focused on improving efficiency (Gooyabadi et al., 2023), reaching new beneficiaries and amplifying the impact of marketing activities (Godefroid et al., 2024), our research illustrates holistically how the AI adoption process is taking place in NPOs starting from operational tasks followed by strategic directions.

However, although the potential of AI in the marketing strategies of NPOs has emerged, our results reveal the main barriers and inhibitors that hinder the full exploitation of these technologies in the third sector. Mainly, the lack of skills and resources to effectively manage AI in marketing strategies poses a significant challenge for organizations (Savastano et al., 2024). Many NPOs struggle with limited access to professionals trained in AI technologies, such as machine learning and predictive analytics, which are essential for leveraging these tools to their full potential. Furthermore, the absence of sufficient financial and infrastructural support hinders the acquisition and implementation of advanced AI systems. The skills gap not only affects the ability to optimize campaigns and target donors efficiently but also limits the capacity to analyze data and derive actionable insights. Without addressing these barriers, NPOs risk falling behind in adopting innovative approaches, reducing their competitiveness and impact in an increasingly data-driven environment.

Academic studies on innovation management in third sector organizations have highlighted that NPOs have made progress in their efforts to adopt emerging technologies, but continue to face various challenges, such as limited resources and a lack of innovation competencies (e.g. Do Adro et al., 2021; Trischler and Li-Ying, 2022). NPOs also encounter difficulties related to the fluidity of resources during their new technologies' adoption process (Nadkarni and Prügl, 2021; Reficco et al., 2021), as well as opportunities and challenges in managing resources in dynamic and evolving contexts (Li et al., 2018; Ponzoa et al., 2023).

The main theoretical contribution of this study is about the advancement of knowledge on NPOs' efforts and obstacles to adopting AI in marketing strategies for creating value. While benefits and opportunities for NPOs associated with the adoption of digital technologies have been highlighted by scholars (e.g. Landoni and Trabucchi, 2024), this is one of the first innovation management studies on third sector organizations which clarifies the role of AI in achieving marketing communications objectives, raising funds and increasing stakeholder engagement. Indeed, although scholars argue that NPOs use digital tools to interact with stakeholders by disseminating information and engaging in conversations to create and sustain an online community of supporters (Chung et al., 2021), the findings of this empirical research focus on the integration of human–AI interactions both in these strategic activities and in the more operational ones. In general terms, our research contributes to the academic debate by emphasizing the importance of the human–machine relationship in the process of adopting AI to make strategic decisions and manage practical activities (Barile et al., 2024; Gaczek et al., 2023). In this vein, our study confirms the recent research by Festa et al. (2025), according to which organizations are aware of the potential benefits of AI applications with specific reference to productivity management but show a lower orientation to AI as a collaborator of human processes. For example, content optimization, integrated management of social media, the collection and analysis of data for predictive purposes and to improve the relationship with stakeholders emerge as key elements guaranteed by AI in the value creation process. In these activities, although AI support is key, human intervention is necessary to align the proposals that come from this new technology with the values, mission and vision of the organizations.

In other words, our findings contribute to the literature by illustrating how the strategic use of AI enhances innovation management practices in NPOs. This includes facilitating automated communication to disseminate information, triggering actions such as donations (Hu and Li, 2020) and fostering long-term stakeholder relationships through increased engagement. Moreover, this study highlights the critical role of managing human–AI interactions as a key component of innovation processes within NPOs (Barile et al., 2024; Simòn et al., 2024). The theoretical implications of this study are related to the integration of AI into the marketing strategies of NPOs, highlighting the crucial role of human–machine interactions in creating value. Its originality lies in the empirical analysis of AI use for operational tasks and in supporting strategic decision-making, offering an innovative perspective on the collaboration between technology and human capital in the third sector. Thus, our analysis expands the academic literature showing that the AI adoption requires theoretical refinement, which accounts for the socio-technical dynamics defining strategic and operational practices. By illustrating how human judgment, organizational values, and mission alignment remain central in leveraging AI for fundraising, stakeholder engagement and marketing communications, this study advances the literature on innovation management strategies in NPOs. In this vein, AI should be seen as a complementary factor and not a substitute in relationships with humans, without undermining employee wellbeing.

The adoption of AI by NPOs has the potential to transform their marketing strategies, their relations with the different stakeholders, and to facilitate some routinized tasks, and consequently to enhance value creation. From our research, it emerged that professionals in the third sector should strategically integrate AI into both long-term planning (strategic level) and day-to-day operations (operational level).

Specifically, NPOs' managers should work on the personalization of fundraising messages through AI tools useful for extending effectiveness and ad hoc communication. Moreover, AI tools, advanced data analysis and machine learning algorithms, could help managers to segment donors and to design engagement cycles based on their interests and past interactions with the organization on one side, but on the other side, the implementation of these tools opens the path to consideration around the ethical and transparent use of data and management of privacy. The success of this strategic process also depends on the interaction with the human component that can correctly read the data, ethically manage it and integrate what emerges with advanced technological tools to positively impact the donor experience.

Third sector managers working on the ability of AI tools to read donors' preferences and interests could improve the empathy and the dynamicity of communication and offer a more fluid donor experience and journey.

In particular, even if AI and algorithms could improve relations with stakeholders and donors, the real value is given by the engagement and active and direct involvement in projects.

While AI can automate and optimize numerous processes, it is crucial to balance technological efficiency with human creativity and empathy, which are key in the third sector. NPO professionals must ensure that AI tools complement, rather than replace, the human touch that is central to nonprofit missions. By addressing these considerations, NPOs can effectively harness AI's transformative potential to enhance their marketing strategies, improve fundraising efforts and build strong relationships with their stakeholders.

Moreover, by leveraging AI for advanced donor behavior analysis and predictive modeling, NPOs can align their marketing campaigns more effectively with organizational goals. In addition, AI facilitates deeper connections with stakeholders through tailored communication, fostering trust and loyalty (Huang and Rust, 2022).

At the operational level, AI offers significant opportunities to optimize resources and increase efficiency (Kedi et al., 2024). Automation of repetitive tasks, such as drafting content, managing social media and scheduling communications, allows staff to focus on mission-critical activities. Furthermore, AI analytics tools enable real-time data utilization, providing insights that allow professionals to dynamically adjust campaigns for maximum impact.

Moreover, our results highlight that AI also plays a key role in enhancing stakeholder engagement (Kumar et al., 2019). AI's ability to personalize interactions through tools like chatbots and sentiment analysis significantly improves the stakeholder experience while fostering stronger relationships (Chandra et al., 2022).

The use of AI tools in the improvement of efficiency is also strictly related to the improvement of financial sustainability of NPOs, usually characterized by limited funds and financial instability, performing better efficiency and increasing performance could improve fundraising activities and campaigns. Moreover, the use of these innovative digital tools could help NPOs in reaching financial goals faster, in improving their social impact and in gaining a competitive advantage.

Despite these benefits, managers must address key barriers to AI adoption in the third sector. For example, the lack of expertise is a significant challenge, requiring a focus on staff training and partnerships with educational institutions or technology providers to bridge the skills gap. Financial constraints also hinder AI implementation; thus, NPO managers should explore partnerships with tech companies, apply for grants or leverage open-source tools to reduce costs. Cultural resistance to adopting new technologies further complicates AI adoption, making it essential for leaders to foster a culture of innovation.

The use of AI tools in NPOs opens the path also to some ethical issues: ethical and responsible use of these innovative and disruptive tools must be prioritized through, for example, the definition of ethical guidelines ensuring that the adoption of new technologies aligns with the mission and the vision of the organizations and is aligned with their values. Moreover, when handling data, NPO managers need to integrate ethical practice in using it, promoting transparency in their utilization to create benefits not only for the organization but also for their beneficiaries and society as a whole.

Policymakers also play a critical role in supporting the adoption of AI within the third sector. For example, they should promote policies that encourage comprehensive digital training programs tailored to the nonprofit workforce, particularly in areas such as AI ethics and tool-specific skills, like using large language models or predictive analytics platforms. Governments and local authorities could fund initiatives similar to the UK's Digital Skills for Charities program or partner with universities and tech companies to create certified online courses accessible to NPO professionals at low or no cost. In addition, the development of specialized skills should be complemented by ensuring equitable access to emerging technologies. This could involve financing the access to AI software licenses or offering cloud credits, so that even small and medium-sized NPOs can experiment with and implement AI solutions without a large investment.

Finally, cross-sector relationships have the potential to serve as a fertile ground for nonprofit managers to voice their concerns regarding the integration of AI in their innovation processes. Such relationships can facilitate the establishment of a shared platform where managers can exchange experiences and identify shared solutions, as well as share a series of best and bad practices from other sectors.

This research focuses on the phenomenon of AI adoption by NPOs to create value across both strategic and operational levels. By highlighting this dual role of AI in supporting long-term planning and enhancing daily activities, our findings highlight the potential of AI in the third sector. More specifically, the research demonstrates that while AI contributes significantly to improving communication, fundraising and stakeholder engagement, its effective implementation depends on the integration with human judgment, creativity and ethical alignment with organizational values. Moreover, this study reveals the barriers that hinder full AI adoption, particularly in terms of limited resources and competencies. In this vein, our research contributes to the academic literature on innovation management in NPOs by emphasizing the importance of managing human–AI interactions to create value. Furthermore, the results of this study provide useful implications for both professionals and policy makers operating in the third sector, revealing how AI can be implemented in their organizations to achieve marketing goals and optimize organizational processes in general, taking also into account the importance of ethical issues and the need for transparency and deliberate strategy development. Strategic planning would be critical and useful to address ethical issues and considerations.

However, the study's reliance on qualitative data collected through semi-structured interviews presents limitations. While this method provides in-depth insights, the findings may lack generalizability across a broader spectrum of NPOs and contexts. Semi-structured interviews, while offering flexibility and depth, have certain limitations that should be acknowledged. The quality and richness of the data collected depend heavily on the skill of the interviewer, including their ability to ask probing questions and manage the flow of conversation. This method can also introduce interviewer bias, as the phrasing of questions or the interviewer's demeanour may influence participant responses. Furthermore, the reliance on participants' information may result in inaccuracies or incomplete accounts, especially if key informants are reluctant to share sensitive details or struggle to articulate their experiences. Moreover, we acknowledge the use of a small sample size, which, while allowing for in-depth exploration, may not have captured the full diversity of perspectives or experiences relevant to the research topic of this paper, which focused on the Italian context. For these reasons, future research could address these limitations by addressing a broader sample of NPOs and by comparing different geographical contexts. Future research could include quantitative approaches, such as large-scale surveys or longitudinal studies, to validate and expand upon the results. Further investigation into the ethical considerations and long-term impacts of AI adoption in the third sector would provide valuable guidance for both practitioners and scholars.

In addition, given the growing interest in the topic and the numerous studies being developed on it, future studies could include a systematic review of the literature that queries the various databases.