Opportunities or challenges? The interplay between artificial intelligence and corporate social responsibility communication

Hua, Xiangzhou; Nurul Ain Mohd Hasan; Feroz De Costa; Qiao, Weihua Article Opportunities or challenges? The interplay between artificial intelligence and corporate social responsibility communication Business Systems Research (BSR) Provided in Cooperation with: IRENET - Society for Advancing Innovation and Research in Economy, Zagreb Suggested Citation: Hua, Xiangzhou; Nurul Ain Mohd Hasan; Feroz De Costa; Qiao, Weihua (2024) : Opportunities or challenges? The interplay between artificial intelligence and corporate social responsibility communication, Business Systems Research (BSR), ISSN 1847-9375, Sciendo, Warsaw, Vol. 15, Iss. 1, pp. 131-157, https://doi.org/10.2478/bsrj-2024-0007 This Version is available at: https://hdl.handle.net/10419/318835 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ 131 Business Systems Research | Vol. 15 No. 1 |2024 Opportunities or Challenges? The Interplay between Artificial Intelligence and Corporate Social Responsibility Communication Xiangzhou Hua, Nurul Ain Mohd Hasan, Feroz De Costa Faculty Of Modern Languages And Communication, Universiti Putra Malaysia, Serdang, Malaysia Weihua Qiao Department of Film & Image, College of Art, CheongJu University, Cheongju, Korea Abstract Background: The rapid development of Artificial Intelligence (AI) offers both opportunities and challenges for its application in Corporate Social Responsibility (CSR) communication. While AI can enhance CSR initiatives, its impact on consumer relations and brand perception remains inconsistent. Objectives: This study aims to explore the academic landscape of AI's role in CSR communication, focusing on publication trends, key authors, research topics, and future directions. Methods/Approach: A bibliometric analysis was conducted on 1,094 articles related to AI and CSR communication, retrieved from the Web of Science database from 2000 to February 2024. Using CiteSpace software, the study mapped research trends by analysing disciplines, countries, institutions, authors, references, and keywords. Results: The United States and China lead in publication output, with key research themes including social media impact, management strategies, and consumer trust. Emerging trends point to the importance of privacy, service quality, and perceived value in AI-driven CSR initiatives. Conclusions: The integration of AI in CSR communication is an evolving field, with significant contributions from social media research and consumer behaviour studies. Future research should address ethical concerns and long-term effects on consumer trust and engagement. Keywords: CSR communication; AI; Consumer; CiteSpace; bibliometric JEL classification: M14, O33 Paper type: Research article Received: Apr 28, 2024 Accepted: Jun 3, 2024 Citation: Hua, X., Hasan, N, A. M., De Costa, F., & Qiao, W. (2024). Opportunities or Challenges? The Interplay between the Artificial Intelligence and Corporate Social Responsibility Communication. Business Systems Research, 15(1), 131-157. DOI: doi.org/10.2478/bsrj-2024-0007 132 Business Systems Research | Vol. 15 No. 1 |2024 Introduction Corporate Social Responsibility (CSR), as a crucial practice concerning the social impact and sustainable development of enterprises, has become one of the focal points in contemporary business. With the rapid development and extensive application of Artificial Intelligence (AI) technology, the application of AI in CSR communication presents new opportunities and challenges for businesses. Through AI technology, enterprises can effectively convey their social responsibility information, enhance their brand image, and better meet the growing expectations of consumers. However, despite some initial achievements in the application of AI in CSR communication, there remain many issues and challenges in this research field that require further exploration. Firstly, research on AI in CSR communication is still in its infancy, especially considering that AI is a widely applicable field, and its application in CSR communication can be both methodological (Wu et al., 2024; Ruhana et al., 2024) and in terms of products and services (KAZEMI et al., 2023; Galiano-Coronil et al., 2024). As of May 2024, a search on the Scopus platform yielded only 88 research articles that simultaneously address both CSR and AI. However, only one relevant article utilising bibliometric analysis has been found so far (Thomas et al., 2024), which also does not delve into the analysis of the impact on and relationship with consumers. Therefore, this study aims to fill this gap by using bibliometric methods. Additionally, most studies remain at the theoretical exploration and case analysis level. Understanding the most pressing concerns of researchers and businesses in this field is crucial for better problem-solving and predicting future research trends. Secondly, there is inconsistency in research findings regarding consumers' attitudes, perceptions, and behavioural responses to AI technology. Some researchers believe that the application of AI makes consumers more receptive to product usage and understand corporate content (Yuan, 2018). However, others argue that consumers subjectively reject AI services, especially when they perceive that AI provides the service, as they feel a lack of warmth and sincerity, leading to opposition to its use. Moreover, a weak point in current research is the lack of long-term effects studies, with research often limited to short-term outcome evaluations and lacking tracking studies on consumers' long-term cognitive and behavioural changes. Finally, a theoretical framework for understanding the impact mechanisms and pathways of AI in CSR communication has yet to be established, lacking systematic organisation and explanation (Cheng et al., 2022). To address these issues, this study will conduct an in-depth analysis using bibliometrics. As a quantitative research method, bibliometrics possesses characteristics of systematicity, objectivity, and scientific rigour, enabling researchers to review existing literature comprehensively and discover patterns and trends therein (Lim et al., 2022; Zhao et al., 2023). CiteSpace, developed by Professor Chaomei Chen, is a Java-based co-citation network visualisation application used to detect and reveal the research frontiers and turning points in knowledge domains (Chen, 2006), primarily applied in bibliometric research (Chen et al., 2014). This study not only employs this software to analyse the publication, country, author, and keyword distribution of research related to the impact of AI on CSR communication with consumers but also makes predictions on trends and summarises the current hotspots and issues faced by the industry and academia in this field based on these analyses. Therefore, this study contributes significantly to the theoretical domain by addressing existing research gaps and potentially advancing further theoretical development. Moreover, it offers valuable practical insights that can aid in the formulation of targeted and effective AI application strategies and practical 133 Business Systems Research | Vol. 15 No. 1 |2024 recommendations for enterprise managers. These insights have the potential to foster corporate social responsibility practices and promote sustainable development. The structure of this study is as follows: The first part introduces the current research status, purpose, and significance of studying the impact of AI on CSR communication with consumers. The second part consists of a literature review on the Application of AI technology in CSR, as well as a review of the literature on Consumer attitudes toward AI for CSR communication. The third part introduces the research methods and presents the specific research process. The fourth part involves data analysis, including results on authors, publications, countries, keywords, and trends. The fifth part proposes conclusions and discussions, covering major findings, contributions, limitations, and future research pathways. Literature review Application of AI technology in CSR Artificial Intelligence (AI) technology has become a cornerstone in the evolution of business practices, with its applications permeating various domains, including Corporate Social Responsibility (Dogru & Keskin,2020). The prevalence of AI in businesses and organisations is highlighted by its ability to mimic human intelligence, enabling machines and models to predict outcomes and provide insights that inform strategic decision-making2. In the realm of CSR, the integration of AI is particularly significant as it aligns business operations with Environmental, Social, and Governance (ESG) factors, addressing the urgent global issues of environmental and social equality (Gore et al., 2023; Atanasov et al., 2023). AI tools such as Natural Language Processing (NLP), data analytics, and machine learning are instrumental in helping businesses understand the impacts of their CSR efforts. By leveraging these technologies, organisations can develop and implement sustainable practices that contribute to their growth while also fulfilling their social and environmental responsibilities1. The transformative power of AI extends beyond CSR, affecting various sectors and industries, from finance to marketing, and reshaping the landscape of business management and operations (Lakhan, 2022). In summary, the application of AI in the CSR domain is a testament to the technology's potential to foster sustainable growth and enhance the efficacy of CSR initiatives. As businesses continue to integrate AI into their economic models, they are better equipped to coordinate actions with ESG factors, ensuring a responsible and sustainable approach to growth (Gore et al., 2023; Lakhan, 2022). Consumer attitudes toward AI for CSR communication Consumer attitudes towards companies using AI technology for Corporate Social Responsibility (CSR) communication are multifaceted and influenced by various factors such as trust, perceived usefulness, and ethical considerations. Trust plays a pivotal role in the acceptance of AI technologies, as demonstrated by a study that found trust to significantly affect the intention to use AI voice assistants, with perceived usefulness and attitude towards the technology acting as mediators (Choung et al., 2023). This indicates that for companies to use AI in CSR communication effectively, they must ensure that their AI systems are perceived as useful and trustworthy. The concept of initial trust, along with perceived risk, has been shown to be critical in the public's acceptance of automated vehicles (Zhang et al., 2019). This suggests that in the context of CSR communication, initial trust in the AI system could be a key determinant of its acceptance. Companies should aim to build this trust by 134 Business Systems Research | Vol. 15 No. 1 |2024 demonstrating the usefulness of the AI system and minimising perceived risks associated with its use. Ethical implications are also a significant concern for consumers when it comes to AI in marketing, which would extend to CSR communication (Gonçalves et al., 2023; Du & Xie, 2021). Perceived risk has a notable impact on attitudes towards AI, ethical concerns, and perceived trust. Therefore, companies must address ethical issues transparently to foster consumer acceptance of AI-driven CSR initiatives. Furthermore, the nature of consumer-AI interactions is influenced by factors such as trust, empathy, attachment, and anthropomorphism (Pop et al., 2023; Wu et al., 2024). Companies should consider these factors when designing AI systems for CSR communication to ensure that the interactions are perceived positively by consumers. Lastly, public opinion towards AI varies by demographic and region, with differences in trust levels observed across gender, socioeconomic status, and geographic location (Zhang, 2021). Companies should be mindful of these differences when using AI for CSR communication to tailor their approach to different segments of the population. In conclusion, for companies to successfully use AI technology for CSR communication, they must build trust and address ethical concerns while considering the demographic and regional differences in attitudes towards AI. Ensuring the AI system is perceived as useful and minimising perceived risks are also crucial for consumer acceptance and trust in this communication approach. Methodology Research approach In this study, we will utilise CiteSpace 6.2.R3 software for literature analysis. CiteSpace is a widely employed tool for academic literature analysis (XIE et al., 2018; Chen, 2000; An & Xing, 2023). CiteSpace has been acknowledged for its macro analysis capabilities, with a focus on literature and periodicals, and its potential for future integration with various visual analysis tools to achieve more accurate results (XIE et al., 2018). Thus, it furnishes a reliable analytical foundation for in-depth exploration of the impact of Artificial Intelligence (AI) on consumer satisfaction within Corporate Social Responsibility (CSR) communication. Several factors were considered when selecting CiteSpace software as our literature analysis tool. Firstly, CiteSpace offers robust visualisation capabilities, facilitating the intuitive presentation of citation relationships, collaborative networks, and thematic evolution among publications. This visualisation capability enables us to comprehend the interconnections among literature more intuitively and swiftly capture research domain hotspots and trends. Secondly, CiteSpace supports network analysis, aiding in the discovery of associative patterns among publications, such as citation relationships and collaborative networks. Through network analysis, we can delve deeper into the connections among literature, identifying latent research topics and pivotal nodes to guide our research direction and focus (An & Xing, 2023). Thirdly, CiteSpace enables time-series analysis of literature data, assisting us in tracking the developmental trajectory and evolutionary trends of research domains, which is crucial for understanding the fluctuation patterns and trends of AI's impact on consumer satisfaction within CSR communication, thereby facilitating a more accurate grasp of the timeliness and forefront of our research. Lastly, CiteSpace supports the export of analysis results in various formats, such as images and tables, facilitating the intuitive presentation of analysis results in research reports or papers. Additionally, the process of using CiteSpace is recordable, ensuring the reproducibility of the research process and 135 Business Systems Research | Vol. 15 No. 1 |2024 enhancing the scientific rigour and credibility of our study (Chen, 2006). In summary, we have chosen CiteSpace software as our literature analysis tool because of its powerful visualisation and analysis functionalities, which can assist us comprehensively and thoroughly explore the impact of AI on consumer satisfaction within CSR communication, thereby providing reliable data support and analytical foundation for our research. Data source and screening This study utilises the Web of Science Core Collection as the data collection platform. Editions selected include the Science Citation Index Expanded (SCI-EXPANDED) from 2000 to February 2024, the Social Sciences Citation Index (SSCI) from 2006 to the present, and the Arts & Humanities Citation Index (AHCI) from 2008 to the present. The data collection process is depicted in Figure 1, with priority given to the titles, abstracts, and keywords of the literature to ensure thoroughness and credibility. The search period spans from January 1, 2000, to February 20, 2024, using the following search criteria: #1 TS="Corporate social responsibility Communication" OR ="CSR Communication" AND #2 TS="AI" OR "Artificial intelligence" AND #3 TS="Consumer", resulting in a total of 1128 articles. Non-journal articles were excluded from the analysis. Other languages are deemed not applicable to this study except English. After screening, 1094 articles meeting the criteria are compiled. Figure 1 Data Collection and Screening Process Source: Author's work These publications cover 158 disciplinary fields, with business, management, environmental studies, and electrical engineering accounting for 65% of the total. Table 1 presents a detailed classification of literature by research disciplines. 136 Business Systems Research | Vol. 15 No. 1 |2024 Table 1 Top 10 research disciplines in this field Ranking Web of Science Categories Record Count % of 1,094 1 Business 382 34.93 2 Management 148 13.54 3 Environmental Studies 89 8.17 4 Engineering Electrical Electronic 77 7.00 5 Environmental Sciences 71 6.50 6 Computer Science Information Systems 69 6.30 7 Green Sustainable Science Technology 66 6.00 8 Communication 55 5.00 9 Telecommunications 49 4.50 10 Computer Science Artificial Intelligence 43 3.93 Source: Author's work Parameter setting and analysis This study employs CiteSpace 6.2.R3 software to detect the frequency or occurrence rate of the keywords "AI" and "corporate social responsibility communication" in the domain of "consumers". The purpose of using this software is to understand the research hotspots and development trends in this field. The time slice is set from January 2000 to February 2024. When setting the parameters, several factors were considered to enhance the effectiveness of the analysis: Firstly, the node types were specified as authors, institutions, countries, keywords, and references. This setting allows for comprehensive consideration of various entities involved in the research domain, including individual contributors, institutional affiliations, geographical distributions, as well as specific terminologies and reference usage. Secondly, the value of "TopN" is set to 50. By selecting the top 50 nodes based on certain criteria (e.g., citation counts), the analysis focuses on the most influential and prominent elements within the network, facilitating a better understanding of key figures and topics in the field. Thirdly, in the g-index calculation, the value of parameter K is set to 10. This adjustment aids in capturing the overall impact and significance of publications by considering both the number of citations received by individual articles and the distribution of these citations among the entire set of articles. Lastly, the network is pruned by selecting "pathfinder" and "pruning merged networks". This trimming process helps refine the network structure, removing redundant or less relevant connections and focusing on core relationships and patterns. These parameter settings in CiteSpace were selected based on previous research and the recommendations of Professor Chaomei Chen, the developer of the software (Chen, 2016), aiming to ensure a comprehensive and in-depth analysis of the research landscape regarding AI and corporate social responsibility communication in the consumer domain. By adopting these parameters, we aim to reveal the development trends and key research topics in this field over the specified period, providing meaningful insights. Bibliometric analysis Publication trend According to the data depicted in Figure 2, there is a stable growth trend in the number of publications concerning artificial intelligence (AI) within the domain of 137 Business Systems Research | Vol. 15 No. 1 |2024 consumer corporate social responsibility communication. As evidenced by the fitted curve, this field gradually garnered attention from the academic community in the early 21st century. Prior to 2012, scholars showed relatively low interest in this area, with an average of only 2.8 articles published annually. However, from 2013 to 2018, there was a substantial increase in the number of relevant articles, the average number of publications per year is about six. In contrast, during the period from 2019 to 2023, the quantity of related publications experienced a significant surge, reaching its peak in 2023. Figure 2 Yearly scientific output from 2000 to February 2024 Source: Author's work The data collected for this analysis concluded on February 17, 2024, and thus does not encompass complete statistical information for the year 2024. Nevertheless, based on the observed trend, we can anticipate a continued increase in the number of publications in this field, aligning with the stable growth trend depicted by the fitted curve formula. To predict the number of publications in 2024 based on the provided data, we can use a trend analysis method. One common approach is linear regression, where we fit a straight line to the data points and use it to estimate future values (Bianchi et al., 1999). Average Growth Rate = [(Final Year Publications / Initial Year Publications) ^ (1 / Number of Years)] - 1 = [(310 / 1)^(1 / 23)] - 1 = 0.2178 or 21.78%. Predicted Publications in 2024 = 2023 Publications * (1 + Average Growth Rate) = 310 * (1 + 0.2178) ≈ 377.8. Therefore, based on the trend analysis using linear regression and the average annual growth rate, the predicted number of publications in 2024 is approximately 378. These figures reflect the increasingly close and significant relationship between artificial intelligence and consumer corporate social responsibility communication, generating widespread attention and discussion within the academic community. 138 Business Systems Research | Vol. 15 No. 1 |2024 Country/region analysis According to the country collaboration network generated by CiteSpace software, the United States, China, and the United Kingdom are the top three countries in terms of the number of published papers in this field (Table 2). Among them, the United States ranks first with 312 papers, accounting for approximately 28.5% of the total literature, demonstrating its strong capabilities and influence in this field. China closely follows, with a total of 228 published papers, accounting for 20.8% of the total, indicating that Chinese scholars also possess a high level of research competence and productivity in this field. Additionally, countries such as Australia, India, and South Korea have gradually increased their research investment in this field and have achieved certain results. In terms of research timeline in each country, the United States not only leads in the number of published papers but also initiates the research as soon as possible (beginning in 2000). Following this, Chinese scholars conducted preliminary research and began publishing related papers. Countries such as the United Kingdom, Spain, Germany, and France conducted relevant research almost before 2010, hence maintaining a stable and continuous increase in publication volume. In contrast, Australia, India, Italy, and South Korea started their research relatively later, with the first batch of research concentrated around 2015. These data reflect differences in the understanding and investment levels of each country regarding the importance of this field. Table 2 The top 10 countries with the highest number of publications Ranking Country Count Year Most cited authors 1 USA 312 2000 Bhattacharya, C B 2 CHINA 228 2002 Kim, S 3 ENGLAND 106 2008 Lu, Vinh Nhat 4 AUSTRALIA 83 2014 Wirtz, J 5 INDIA 66 2013 Pillai, R 6 SOUTH KOREA 66 2015 Sohn, K 7 SPAIN 62 2009 Currás-Pérez, R 8 ITALY 58 2013 Pesapane, F 9 GERMANY 58 2009 Du, S 10 FRANCE 52 2008 Parguel, B Source: Author's work CiteSpace employs several centrality metrics to assess the importance of documents or authors, with one of the most common being "Betweenness Centrality." The betweenness centrality is defined for each node in a network. It measures how likely an arbitrary shortest path in the network will go through the node. The calculation of Betweenness Centrality is to divide the sum for each node by the total number of possible node pairs (excluding the node itself) to obtain its Betweenness Centrality value. If a node has a high mediation centrality in the network, it means that it plays a key mediating role in information dissemination and cooperation (Chen, 2000). In addition to numerical advantages, Lithuania (0.73), Romania (0.57), Pakistan (0.45), France (0.44), and Germany (0.38) have the highest centrality among countries (Table 3), this shows that the above-mentioned countries have had a certain influence and made outstanding contributions in this field. It is worth noting that conducting scientific and technological cooperation in different regions and backgrounds will better promote global technological innovation and progress. 145 Business Systems Research | Vol. 15 No. 1 |2024 consumer behaviour. These three directions are currently hot topics in this field and are of significant practical importance. Firstly, research related to CSR communication primarily focuses on clusters 0, 1, 2, 5, and 6. The form of CSR communication is profoundly influenced by industry and institutional pressures and organisational characteristics, challenging the universality of CSR initiatives (O'Connor & Gronewold, 2013). Regarding the application and performance of AI in CSR communication, research is mostly conducted on social media platforms. The dissemination and methods of CSR on social media are concerns for researchers and businesses. Abitbol and Lee (2017) explored how companies strategically use dedicated pages on Facebook to engage stakeholders and found that alignment with core business themes can effectively attract stakeholders. Meanwhile, an analysis of the communication strategies of Chinese and Japanese companies on Facebook and Twitter revealed that Chinese companies prefer a diversion strategy and demonstrate faster and stronger handling capabilities of online complaints (Yuan, 2018). A survey found that consumers' overall response to AI influencers on Instagram is less trustworthy than traditional influencers (Sands et al., 2022), leading researchers to caution brands against hastily replacing human influencers with AI counterparts. Text-based chatbots are considered innovative but perform poorly on social media platforms; frequent failures lead to consumer resistance (Cheng et al., 2022). Second, research on artificial intelligence (AI) and its applications and learning primarily focuses on clusters #3, #4, #8, and #9. Intelligent AI devices have become a common presence in the commercial sector, providing a wide range of services from the medical industry to the hospitality sector. From an organisational perspective, AI devices offer several advantages over humans, enabling faster and more accurate execution of certain tasks while also being more cost-effective (Pelau et al., 2021). Consequently, the application and study of AI have become the focal point of researchers and businesses in recent years. Chhetri (2024) proposes that the integration of AI and machine learning technologies in the food industry, particularly in food quality control and safety, not only enhances efficiency and minimises risks but also ensures regulatory compliance, heralding a new era of personalised nutrition, autonomous monitoring, and global collaboration, marking a paradigm shift in the field of food quality control and safety (Chhetri, 2024). However, the greatest challenge facing the investment and proliferation of AI in enterprises and markets is how to gain better acceptance from stakeholders, with many scholars offering constructive suggestions. Currently, the impact of AI on stakeholders, especially consumers, is not very pronounced, and there are even cases of consumer resistance to AI. Consumer engagement through AI (such as voice assistants) does not directly influence future purchasing intentions (McLean et al., 2021). However, it has been found that when humanoid AI devices can demonstrate empathy and interaction similar to human consumers, they are more likely to be accepted (Pelau et al., 2021). The anthropomorphised attributes of perceived warmth and perceived capability positively affect consumers' perceived trust in chatbots, while communication delay has a negative impact (Cheng et al., 2022). Furthermore, researchers have found that consumers' acceptance of highly innovative products with very low practical value (such as AI-based smart products) is more influenced by technological interest than utilitarian aspects. Therefore, when facing highly innovative products, consumers are not inclined to "pay the bill" solely for the sake of pursuing "low prices" (Sohn & Kwon, 2020). Yen and Chiang (2020) found that chatbots in AI enable businesses to enhance customer experiences and meet expectations through real-time interactions in e-commerce environments. Thus, 146 Business Systems Research | Vol. 15 No. 1 |2024 factors influencing consumers' trust in chatbots are crucial. Credibility, capability, anthropomorphism, social presence, and informativeness affect consumers' trust in chatbots, thereby influencing purchase intentions. Finally, research on consumer behaviour has long been favoured by researchers because consumer behaviour determines a company's profits and reputation. Through the organisation and summarisation of a large body of relevant studies, it has been found that consumer behaviour is influenced by CSR communication and technological applications. Consumer behaviour is the study of how individuals, groups, and organisations choose, acquire, use, and provide ideas, goods, and services to meet their requirements and needs. It refers to the behaviour of customers in the market and its underlying reasons. Customer Behaviour Modelling (CBM) in the field of AI is a frequently mentioned technical concept by researchers. It is constructing a mathematical structure to capture typical behaviours observed in certain customer groups so that similar behaviours can be predicted under comparable conditions. Meanwhile, scholars have explicitly stated a positive correlation between CSR communication and consumer behaviour. In the banking sector, the use of technology has a positive and direct impact on service quality, customer satisfaction, and consumer purchasing behaviour (Omoge et al., 2022). Corporate social responsibility communication contributes to increasing consumers' likelihood of purchase and admiration for banks (Gupta et al., 2021). Including Generation Y, corporate social responsibility support plays a central role in regulating purchase intentions (Luger et al., 2022). In addition to research on traditional CSR communication and consumer behaviour, scholars have studied the impact of new technologies appearing in social media on consumer behaviour. Laszkiewicz and & Kalinska‐Kula (2023) conducted research on the influence of computer-generated avatars and virtual influencers (VIs), including those using artificial intelligence (AI) and machine learning, on consumer attitudes and behaviour in social media, revealing the mechanisms of VI attractiveness and their role in shaping consumer attitudes and behaviour. Research trend analysis Time zone analysis As a visual representation of data, a time zone map not only clearly depicts the distribution and changes of keywords in different periods within specific topics or fields but also aids in better understanding the evolving hot topics in the field of CSR communication. Through in-depth analysis of time zone maps, rich information can be extracted, such as the frequency of use of specific keywords or phrases during certain periods, the relationships between related topics, and trend changes. Regarding the impact of AI on consumer satisfaction in CSR communication, the time zone view can clearly display the usage of popular phrases and keywords from 2000 to 2024, spanning approximately 24 years (Figure 7). These data reflect the challenges, opportunities, and evolving trends related to the application of AI technology in CSR communication. By delving into and analysing these data, we can more accurately grasp the challenges faced by AI technology in different stages of CSR communication and provide a reference for future development of rational and effective communication strategies. Therefore, the time zone map has become an important tool for understanding the dynamics of the research field. 147 Business Systems Research | Vol. 15 No. 1 |2024 Figure 7 Time zone knowledge map Source: Author's work Although data collection results show that research on the application of AI in CSR communication and its impact on consumers emerged as early as 2000, analysis reveals that it was not primarily within the disciplines of marketing, economics, and management but rather concentrated in medical and biological-related disciplines. Furthermore, research in this field has not been extensive, with scholars only beginning to study and organise it in 2017 systematically. Over these 7 years, the research can be categorised into three distinct phases: 2017-2019, 2020-2022, and 2023. However, these phases cannot be accurately classified according to the conventional stages of "initiation," "development," and "maturity" because AI in CSR communication and consumer domains is a relatively novel concept. Instead, over these 7 years, the field more closely resembles an initiation phase. As researchers delve deeper, as businesses apply AI in practice and as consumers and markets react, various challenges and issues arise alongside the convenience of AI in CSR communication with consumers. Therefore, the author primarily divides these three periods based on the differing focal points of research questions during each period. Between 2017 and 2019, research in this field primarily focused on the impact of corporate social responsibility communication on consumers, who often utilised social media platforms for application. However, researchers began to shift away from traditional questionnaire analysis methods towards using media data or more specific analytical approaches, laying a solid foundation for subsequent AI learning and improvement. Etter et al. (2018) suggested using social media data and sentiment analysis to study citizens' emotionally based responses to organisational behaviour. Some scholars provided practical recommendations for companies engaging in CSR communication through social media. They suggested that companies could use hashtags to make thematic information clear and leverage discussions within existing social movements to resonate with consumers, thereby conveying the theme of corporate social responsibility (Saxton et al., 2019). Additionally, frequent tweeting 148 Business Systems Research | Vol. 15 No. 1 |2024 about corporate social responsibility on social media platforms was found to garner higher user approval (Araujo & Kollat, 2018). From 2020 to 2022, research in this field primarily focused on two aspects: exploring AI technology and discussing consumer trust in AI. Regarding the exploration of AI technology, this included machine learning, voice assistants, text recognition, and more. Huang and Rust (2020) found a three-stage framework for strategic marketing planning, integrating various AI advantages: mechanical AI for automating repetitive marketing functions and activities, thinking AI for processing data to make decisions, and emotional AI for analysing interactions and human perceptions. Buhalis and Moldavska (2022) found through research on interactions supported by Voice Assistants (VAs) between hotels and guests in the hotel industry that although VA technology is still in its infancy, virtual assistants are increasingly becoming digital assistants, helping hotels improve customer service, expand operational capabilities, and reduce costs. As for the issue of consumer trust in the application of AI technology in CSR communication, it appears to be one of the concerns and focuses for researchers and businesses. Artificial Intelligence (AI) voice assistants (VAs) continue to be popular among consumers, and while technological means do influence consumers' brand engagement, concerns about trust issues that may arise after AI usage can negatively impact brand engagement behaviour (McLean et al., 2021). Uysal's interviews and on-site research with Artificial Intelligence Assistants (AIAs) users indicated that the anthropomorphism of AIAs may threaten users' identity, thereby depriving them of their power, triggering data privacy issues, and ultimately harming their well-being. These harmful effects are particularly evident in intimate, long-term relationships (Uysal et al., 2022). In the past year of 2023, researchers have focused more on addressing trust issues and improving how AI serves CSR communication with consumers. Particularly, addressing ethical issues mainly centred around trust problems has drawn attention from practitioners and scholars. Liu et al. (2021) highlighted the need for scrutiny of ethical issues associated with the use of artificial intelligence. This scrutiny can help organisations gain trust, minimise privacy infringements, and ultimately promote responsible success in operating AI on social media platforms. Meanwhile, an increasing number of scholars are utilising technology acceptance models (TAM) to explain the relationship between consumer perception and AI usage. Iancu and Iancu (2023) found in their study involving elderly consumers in the healthcare industry using AI-related systems that perceived ease of use was explained by effort, ability, and perceived external control when interacting with chatbots, while perceived usefulness was supported by perceived ease of use and subjective norms. Additionally, if individuals perceive the interaction as useful and receive approval from others, they may further utilise chatbots (behavioural intention). This insight suggests that in the design of future AI products and systems, ease of use should be the primary consideration, following responsible AI principles, and effective promotion of AI products and systems should also be conducted to influence consumer perceptions positively. Previous research indicates that consumers perceive nonhuman (compared to human) agents differently in many aspects (e.g., lacking emotional capability), leading to various behavioural changes when interacting with them. Kim et al. (2022) found that users tend to engage in unethical consumer behaviour more when interacting with non-human (as opposed to human) agents (e.g., artificial intelligence and robots) because the anticipated guilt is reduced. Furthermore, he proposed the moderating effect of anthropomorphism, suggesting that imparting non-human agents with human-like characteristics can reduce 149 Business Systems Research | Vol. 15 No. 1 |2024 unethical behaviour. Therefore, anthropomorphism in AI regulation can not only increase user trust and experience but also effectively reduce user unethical behaviour. Keywords' burst analysis Keyword burst analysis is an important method based on big data, which helps the scientific community to understand hot issues in a timely manner by analysing the frequency changes of specific keywords in a particular field. This method not only reveals valuable information about research frontiers and trend predictions but also provides valuable data references for decision-makers. The intensity and duration of keyword bursts over the past 24 years, as shown in Figure 8, reflect the achievements of this method in practical applications. Through in-depth analysis of these data, we can better understand issues that suddenly attract attention in certain fields and provide guiding significance for future related research. Furthermore, with the continuous advancement of technologies such as artificial intelligence and machine learning, keyword burst analysis will also usher in new development opportunities. Figure 8 Keywords' burst analysis Source: Author's work In CiteSpace operations, the period should be set to at least one year, and parameters such as gamma 1.0, f(x) = alpha * e - alpha x (where alpha 1 / alpha 0 = 2.0, alpha I/alpha I - 1 = 2.0) should be used for analysis. The red part indicates the period during the outbreak of the epidemic, during which the keyword "management" became a topic of continuous discussion and attention. In addition, over the past 24 years, "business" and "social responsibility" have also been two keywords with long durations and high levels of attention. These data not only reflect the enduring high involvement of people in issues such as business economy and social responsibility but also help us better understand the changes and evolution of different thematic topics within various fields over time. 150 Business Systems Research | Vol. 15 No. 1 |2024 Analysis of cited reference burst According to the data from Figure 9, citations related to artificial intelligence in the field of consumer social responsibility communication show a steady increase year by year, which indicates that research on artificial intelligence in consumer social responsibility has garnered widespread attention from both academia and industry. The primary period is set to at least 2 years in the Citespace. This decision was made because journals published within a one-year period are numerous and widely distributed, allowing for a better reflection of the development of artificial intelligence and consumer social responsibility communication. From the graph, it can be observed that articles such as Cho et al. (2017), Crane et al. (2016), and Etter et al. (2018) have experienced significant citation bursts, indicating that many scholars cited these journals during that period, and these articles hold significant importance in the field. This further confirms that the intersection of artificial intelligence and consumer social responsibility communication is becoming a focal point of attention for both academia and industry. Figure 9 Cited references burst Source: Author's work There is an increasing acknowledgement that the design, implementation, and effectiveness of CSR initiatives are not merely internal management concerns; rather, they necessitate robust communication and collaboration with stakeholders. These stakeholders include employees, customers, suppliers, investors, government agencies, and local communities, among others. They have direct or indirect impacts on the actions and decisions of the corporation. Therefore, establishing good communication channels with them and actively listening to their opinions and feedback are crucial for the success of CSR initiatives. Only through effective communication with stakeholders can their needs and expectations be better understood and considered in CSR planning, ensuring that the design, implementation, and success of CSR initiatives align with the principles of social responsibility and have a positive impact (Crane & Glozer, 2016). The communication between businesses and users regarding CSR has been a topic of discussion for nearly two decades. However, even among the 'World's Most Admired Companies' listed by 151 Business Systems Research | Vol. 15 No. 1 |2024 Fortune magazine, the frequency of non-CSR messages conveyed in social media posts by companies exceeds that of CSR messages. When conveying CSR activities, companies tend to use informative strategies rather than interactive strategies, and they include activities targeting internal audiences more frequently than those targeting external audiences (Cho et al., 2017). Effective CSR communication is not something that can be achieved overnight; it requires time and strategy, which is challenging for business managers. For the application of new technologies such as AI, as early as 2016, Doorn emphasised in their research the ability of technology to attract customers at the societal level, introducing a new concept called Automated Social Presence (ASP), which refers to the extent to which technology makes customers feel the presence of another societal entity. Based on this concept, Doorn also suggested that in response to the impact of new technologies on consumers, companies should focus on the application of technology in relation to customer relationships, the tendency for technological anthropomorphism, and technological readiness (van Doorn et al., 2017), which aligns with the previous discussion on the timeline of keywords and reflects the focus and trends of research in recent years and the future. Conclusions Contributions The purpose of this study is to provide an overview of the application of AI in the field of CSR communication and its relationship with consumers. Through bibliometric analysis of literature, we can identify patterns, and characteristics within relevant thematic research, highlighting current areas and potential future directions. CiteSpace (V6.2.R3) is a widely used tool for this purpose; therefore, we selected it for our analysis. We searched the WOSCC database using keywords "AI," "Corporate social responsibility communication," and "Consumer," applying specific criteria that resulted in 1094 articles published over a span of 24 years (from 2000 to February 2024). After categorising the 158 disciplines, we identified the top five key disciplines as business, management, environmental studies, engineering, electrical engineering, and environmental sciences. Firstly, in terms of publication volume, before 2012, scholars showed limited interest in this field, with very few focusing on the application of AI in CSR communication and its impact on consumers. With the introduction of the AI concept, from 2013 to 2018, there was a noticeable increase in the number of articles published. In contrast, from 2019 to 2023, the research related to this topic saw a sharp increase, with not only theoretical studies but also more scholars analysing problems encountered by businesses in the market and providing relevant recommendations through experimental methods, which indicates that research in this field has received increasingly widespread attention in the academic community. At the same time, we expect the number of publications in 2024 to be approximately 378. Secondly, the authors conducted an in-depth analysis and identification of key contributors and collaborators in the field. Cheng Yang as one of the most influential authors in the field, with the publication of 8 articles. Following closely are Lu Yaobin, Dong Tianqing, and Cui Yuanyuan (Gina), each with 6 articles published in the field. Moreover, besides individual contributions, prominent institutions such as the University of London, Copenhagen Business School, and Auckland University of Technology have also made significant literature contributions in this domain. In country/region analysis, the United States leads in research publications, followed by China and the United Kingdom, respectively, reflects the strong research 152 Business Systems Research | Vol. 15 No. 1 |2024 capabilities of these countries in the field and their high level of attention and investment in related issues. It is worth noting that there are many other countries or regions globally that actively participate and make significant contributions, which may become even more active and prominent players in the future. Thirdly, the analysis of keyword distribution in the field of AI application in CSR communication and relationship with consumers reveals that "social media," "impact," and "management" are the top three contributing keywords. In order to make the analysis clearer, the software generated 12 clusters with different names. For example: #0 social media, #1 corporate social responsibility, #2 corporate communication, #3 artificial intelligence, #4 deep learning, #5 CSR communication, #6 annual reports, #7 customer satisfaction, #8 service experience, #9 internet of things, #11 consumer behaviour. After clustering these 12 groups, the authors classified them into three research directions based on their content, namely research related to corporate CSR communication, research on AI and its applications and learning, and research on consumer behaviour. Fourthly, when classified according to the timeline of keywords, it was found that the research in this field is relatively recent, with scholars only beginning to research and organise it from 2017 systematically. Over these 7 years, the research can be divided into three stages: 2017-2019, 2020-2022, and 2023. Overall, during these 7 years, this field remains in the developmental stage. Particularly, with the increasing adoption and popularisation of new technologies such as AI, more and more researchers are focusing on addressing trust issues and improving AI's service to CSR communication for consumers. Ethical issues, particularly those related to trust, have garnered attention from practitioners and scholars alike. Finally, in burst word analysis, "management" has been the longest burst word in the past 24 years. Other keywords with lasting impact include "business" and "social responsibility". As CSR communication and consumer relations are inherently part of the disciplines of management and business, many scholars use disciplinary classification as a key criterion. 'Privacy', 'Service', 'Perceived value', and 'Online' are newly emergent keywords this year, indicating potential directions for future research. It also proves that issues of trust and ethics have always been issues that urgently need to be addressed and paid attention to in this field(Gonçalves et al., 2023; Du & Xie, 2021). At the same time, we found through research that some scholars have given relevant and effective suggestions (Buhalis & Moldavska, 2022; Etter et al., 2016). Researchers and businesses increasingly focus on the impact of businesses on consumers when utilising artificial intelligence for CSR communication. Particularly, there is a growing emphasis on using new technologies such as artificial intelligence to improve service, enhance consumer perceived value, better protect consumer privacy, and consequently alleviate consumer concerns about artificial intelligence. Through visual analysis of knowledge graphs, we can gain insights into the application of artificial intelligence in corporate social responsibility communication and consumer relations, which helps to understand the knowledge structure of this field comprehensively and provides valuable references for related research. Firstly, we can summarise the trends and distribution of contributions in the articles to demonstrate that the application of artificial intelligence in corporate social responsibility communication is a continuous and highly valued research direction. This is significant as it impacts consumers. Secondly, we can delineate the foremost contributors—namely, the authors, institutions, and nations—that significantly impact this domain. Furthermore, we shall delve into an intricate analysis of keyword clusters and explore the most frequently cited articles. This information will help us better understand the current research status. Thirdly, during the visual analysis of knowledge 153 Business Systems Research | Vol. 15 No. 1 |2024 graphs, we can uncover the topics and themes of great interest, guiding future related research. Fourthly, after delving into the evolution of research on the impact of artificial intelligence on corporate social responsibility communication to consumers, it is necessary to classify and elaborate on it according to content. Additionally, analysing and summarising future research directions based on highly cited articles will assist future scholars in finding relevant statistical data and analytical materials, thus promoting further development in this field. In practical terms, firstly, scholars provide more reflections on corporate social responsibility communication strategies, encouraging companies to leverage social media to increase the frequency and content of CSR communication. Simultaneously, regarding the development and application of technologies like AI, emphasis should be placed on AI's simulation of human emotions. Moreover, when companies utilise AI-related products and services targeting consumers, ensuring consumer convenience should be a priority. Additionally, adequate pre-promotion efforts are essential, potentially fostering greater consumer acceptance of AI-related services and products. Regarding ethical concerns and privacy issues, companies should adhere to responsible AI principles to better serve consumers and the market. Limitations and future studies Research on the impact of artificial intelligence on CSR communication with consumers still has certain limitations. Apart from the limitations of bibliometric analysis, there are other aspects to consider. Firstly, due to the continuous development and changes in artificial intelligence technology itself, research results may vary at different time points. Secondly, during the data collection process, factors such as sample selection bias and data quality may affect the accuracy and reliability of research conclusions. Additionally, relevant literature may be omitted when searching for keywords. For example, not all fields or specialised terms related to corporate social responsibility communication and artificial intelligence are covered. Therefore, in future research, it may be beneficial to use broader search strategies to obtain more relevant literature and consider introducing other databases for comparative analysis to improve data coverage. The application of AI in CSR communication involves multiple fields, such as ethics, law, and business, thus necessitating interdisciplinary research and collaboration. For instance, in the ethical domain, artificial intelligence may impact consumer privacy, while in the legal realm, it requires clarity on relevant responsibilities and regulations. Moreover, businesses need to enhance their communication skills and methods to convey their social responsibility better and enhance consumer awareness of their actions. Additionally, at the consumer level, there are divergent attitudes toward the use of artificial intelligence in CSR communication. Some consumers are concerned that artificial intelligence may lead to job losses or replace traditional service methods, while others value the convenience and efficiency it brings. Therefore, when engaging in CSR communication, it is crucial to consider the needs and concerns of different groups and seek more effective communication methods. Addressing these issues, academia and businesses can strengthen collaborative research efforts to explore how new technologies, such as artificial intelligence, can improve CSR communication and alleviate consumer concerns about ethical implications. References 1. Abitbol, A., & Lee, S. Y. (2017). Messages on CSR-dedicated Facebook pages: What works and what doesn't. Public Relations Review, 43(4), 796–808. https://doi.org/10.1016/j.pubrev.2017.05.002 154 Business Systems Research | Vol. 15 No. 1 |2024 2. An, S., & Xing, M. (2023). The application of CITESPACE Software in document visual analysis and writing. Third International Conference on Artificial Intelligence, Virtual Reality, and Visualization (AIVRV 2023), 41. https://doi.org/10.1117/12.3011365 3. Araujo, T., & Kollat, J. (2018). Communicating effectively about CSR on Twitter. Internet Research, 28(2), 419–43. https://doi.org/10.1108/intr-04-2017-0172 4. Atanasov, A., Chipriyanova, G., & Krasteva-Hristova, R. (2023). Integration of digital technologies in Corporate Social Responsibility (CSR) ACTIVITIES: A systematic literature review and Bibliometric analysis. Journal of Risk and Financial Management, 16(8), 373. https://doi.org/10.3390/jrfm16080373 5. Bianchi, M., Boyle, M., & Hollingsworth, D. (1999). A comparison of methods for trend estimation. Applied Economics Letters, 6(2), 103–109. https://doi.org/10.1080/135048599353726 6. Buhalis, D., & Moldavska, I. (2022). Voice assistants in hospitality: using artificial intelligence for customer service. Journal of Hospitality and Tourism Technology, 13(3), 386-403. https://doi.org/10.1108/jhtt-03-2021-0104 7. Chen, C. (2016). CiteSpace: A practical guide for mapping scientific literature. Nova Science Publishers. 8. Chen, C. (2000). CiteSpace Home. CiteSpace. https://citespace.podia.com/ 9. Chen, C. (2006). CiteSpace II: Detecting and visualizing emerging trends and transient patterns in scientific literature. Journal of the American Society for Information Science and Technology, 57(3), 359-377. https://doi.org/10.1002/asi.20317 10. Chen, C., Dubin, R., & Kim, M. C. (2014). Emerging trends and new developments in Regenerative Medicine: A scientometric update (2000 – 2014). Expert Opinion on Biological Therapy, 14(9), 1295–1317. https://doi.org/10.1517/14712598.2014.920813 11. Cheng, X., Zhang, X., Cohen, J., & Mou, J. (2022). Human vs. AI: Understanding the impact of anthropomorphism on consumer response to Chatbots from the perspective of trust and relationship norms. Information Processing & Management, 59(3), 102940. https://doi.org/10.1016/j.ipm.2022.102940 12. Chhetri, K. B. (2024). Applications of Artificial Intelligence and Machine Learning in Food Quality Control and Safety Assessment. Food Engineering Reviews, 16(1), 1-21. https://doi.org/10.1007/s12393-023-09363-1 13. Cho, M., Furey, L. D., & Mohr, T. (2017). Communicating Corporate Social Responsibility on Social Media: Strategies, Stakeholders, and Public Engagement on Corporate Facebook. Business and Professional Communication Quarterly, 80(1), 52-69. https://doi.org/10.1177/2329490616663708 14. Choung, H., David, P., & Ross, A. (2023). Trust in AI and Its Role in the Acceptance of AI Technologies. International Journal of Human-Computer Interaction, 39(9), 1727-1739. https://doi.org/10.1080/10447318.2022.2050543 15. Crane, A., & Glozer, S. (2016). Researching Corporate Social Responsibility Communication: Themes, Opportunities and Challenges. Journal of Management Studies, 53(7), 1223-1252. https://doi.org/10.1111/joms.12196 16. Dogru, A. K., & Keskin, B. B. (2020). AI in operations management: applications, challenges and opportunities. Journal of Data, Information and Management, 2(2), 6774. https://doi.org/10.1007/s42488-020-00023-1 17. Du, S., & Xie, C. (2021). Paradoxes of artificial intelligence in consumer markets: Ethical challenges and opportunities. Journal of Business Research, 129, 961-974. https://doi.org/10.1016/j.jbusres.2020.08.024 18. Etter, M., Colleoni, E., Illia, L., Meggiorin, K., & D'Eugenio, A. (2018). Measuring Organizational Legitimacy in Social Media: Assessing Citizens' Judgments With Sentiment Analysis. Business & Society, 57(1), 60-97. https://doi.org/10.1177/0007650316683926 19. Galiano-Coronil, A., Aguirre Montero, A., López Sánchez, J. A., & Díaz Ortega, R. (2024). Exploring social responsibility, social marketing and happiness using artificial intelligence, automated text analysis and correspondence analysis. Management Decision, 62(2), 549-574. https://doi.org/10.1108/md-01-2023-0099