Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Marketing with ChatGPT: Navigating the Ethical Terrain of GPT-Based Chatbot Technology

Marketing with ChatGPT: Navigating the Ethical Terrain of GPT-Based Chatbot Technology Commentary Marketing with ChatGPT: Navigating the Ethical Terrain of GPT-Based Chatbot Technology 1, ,†,‡, 2,‡ Pablo Rivas * and Liang Zhao Department of Computer Science, Baylor University, Waco, TX 76798, USA Department of Marketing, St. Ambrose University, Davenport, IA 52803, USA * Correspondence: pablo_rivas@baylor.edu; Tel.: +1-254-710-3385 † Current address: Center for Standards and Ethics in Artificial Intelligence, Waco, TX 76798, USA. ‡ These authors contributed equally to this work. Abstract: ChatGPT is an AI-powered chatbot platform that enables human users to converse with machines. It utilizes natural language processing and machine learning algorithms, transforming how people interact with AI technology. ChatGPT offers significant advantages over previous similar tools, and its potential for application in various fields has generated attention and anticipation. However, some experts are wary of ChatGPT, citing ethical implications. Therefore, this paper shows that ChatGPT has significant potential to transform marketing and shape its future if certain ethical considerations are taken into account. First, we argue that ChatGPT-based tools can help marketers create content faster and potentially with quality similar to human content creators. It can also assist marketers in conducting more efficient research and understanding customers better, automating customer service, and improving efficiency. Then we discuss ethical implications and potential risks for marketers, consumers, and other stakeholders, that are essential for ChatGPT-based marketing; doing so can help revolutionize marketing while avoiding potential harm to stakeholders. Keywords: marketing; ChatGPT; GPT technology; Chatbot; ethics; societal implications 1. Introduction to ChatGPT Citation: Rivas, P.; Zhao, L. On 30 November 2022, OpenAI unveiled a chatbot named ChatGPT, quickly draw- Marketing with ChatGPT: ing the attention of artificial intelligence (AI) researchers and scholars who tested the Navigating the Ethical Terrain of technology over several hours and days to follow. The release of ChatGPT also sparked GPT-Based Chatbot Technology. AI widespread public interest, with individuals across the globe eager to experience the in- 2023, 4, 375–384. https://doi.org/ novation and assess its capabilities. In regions where ChatGPT is not yet available, such 10.3390/ai4020019 as China, the acquisition of a ChatGPT login has become a highly coveted commodity, fueling the growth of an illicit market in which more than 600 vendors peddle logins Academic Editor: Rafał Drezewski ˙ for a nominal fee [1]. Designed to generate detailed textual responses and articulate an- Received: 18 February 2023 swers to any inquiry [2], ChatGPT has garnered attention for its potential application in Revised: 1 April 2023 the field of engineering geology [3]. One study demonstrated that 70% of users favored Accepted: 4 April 2023 ChatGPT-powered conversational interfaces over conventional methods, citing conve- Published: 10 April 2023 nience, efficiency, and personalization [4]. However, some experts are wary of the tool, with a large AI conference banning its use in conference papers [5]. Additionally, the journal Science has stated that no text generated by ChatGPT or any other AI tool can be used in a paper [6]; Elsevier has established a policy on AI-assisted technologies in scientific Copyright: © 2023 by the authors. writing that requires authors to disclose their use, and not list or cite these tools as authors Licensee MDPI, Basel, Switzerland. or co-authors, limiting the use of AI for enhancing the readability and language of the This article is an open access article work [7]. distributed under the terms and ChatGPT is an AI-powered chatbot platform that utilizes natural language process- conditions of the Creative Commons ing and machine learning algorithms to enable human users to converse with machines, Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ thereby transforming how people interact with AI technology [8]. Before this advancement, 4.0/). most commercialized AI tools were conventional bots or voice assistants with limited AI 2023, 4, 375–384. https://doi.org/10.3390/ai4020019 https://www.mdpi.com/journal/ai AI 2023, 4 376 applications. In contrast, ChatGPT offers numerous advantages over previous AI tools and holds significant potential for application in various fields, including customer service, edu- cation, health care, finance, entertainment, creative writing, digital marketing, e-commerce, and many others [9]. For example, e-commerce companies could leverage ChatGPT to provide automated customer service and order fulfillment services round-the-clock, re- ducing personnel resource costs. Similarly, educators could harness ChatGPT to develop individualized study plans tailored to each student’s interests and progress and to deliver instantaneous feedback through automated grading and virtual assistance in answering questions. In the healthcare sector, healthcare professionals could use ChatGPT to triage symptoms, conduct initial consultations, and even diagnose conditions. The customized and empathetic nature of these conversations may promote a more patient-centered ap- proach and alleviate the anxiety and social pressure that patients may experience when discussing certain health conditions, such as mental illness, with an actual human being [8]. A content marketing company recently conducted an experiment utilizing ChatGPT to create promotional Twitter posts advertising an upcoming webinar on whether robots will eventually replace human labor in content marketing [10]. Here are examples of the two generated Twitter posts: Can you tell the difference between human-written and AI-generated content? Put your skills to the test in our upcoming webinar “ChatGPT vs. Content Marketing” on 15 Feb at 2 p.m. EST. Register now! #AI #conTentmarketing #webinar. The future of marketing is here and it’s all about AI. Don’t miss our webinar “Chat GPT vs. Content Marketing” on Wed. 15 Feb at 2 p.m. EST to learn how to harness the power of AI tools like ChatGPT for your content marketing. Register now! #AI #contentmarketing #webinar. The AI-generated posts were completed much more efficiently than human-generated posts, and even suggested hashtags for users. The list of potential benefits goes on [11]. While these benefits are promising, the tool’s limitations and potential ethical implications must also be considered. Despite the excitement surrounding ChatGPT and its numerous applications, some individuals have expressed ethical concerns regarding this latest inno- vation [12,13]. More specifically, there are known risks associated with AI-based chatbots; for ChatGPT, these mainly include the following [11,14]. ChatGPT can collect and analyze large amounts of personal data, leading to privacy violations if the data is not adequately protected [15,16]. Bias can also be perpetuated and amplified by ChatGPT if it is not ad- equately designed and tested [17,18]. In addition to these risks, ChatGPT can lead to job losses as it automates specific tasks previously done by humans [19]. As ChatGPT becomes more sophisticated, companies may become too reliant on it, leaving them vulnerable if the tool fails or if the data used to train it is inaccurate [20]. Furthermore, the tool has been perceived as a new way to cheat, making the AI write the work and submit it as your own for credit, which generally constitutes plagiarism [12]. Fur- thermore, training ChatGPT has caused severe mental stress on its content curators [21,22]. ChatGPT can generate incorrect, incomplete, misleading, biased, harmful information, instructions, or content [23]. In addition, it sometimes generates fake outputs that are not real, which can erode the trust in citing science [24]. Finally, we emphasize that ChatGPT does not have a human oversight system to review and vet all its outputs, which can lead to ethical violations [25]. It is also important to note that ChatGPT is not unique in its potential for ethical implications. Instead, it is a representative example of large-scale pre-trained language models (PLMs), a rapidly developing field in Natural Language Processing (NLP) [26]. To illustrate the potential ethical implications of large-scale PLMs, we chose to focus on ChatGPT as a case study, examining its specific characteristics and design to draw broader implications for large-scale PLMs, although our focus is on marketing. In the following section, we will explore in detail the implications of ChatGPT in the field of marketing and how it can be used to address some of the ethical concerns while maximizing its benefits. AI 2023, 4 377 This ethical analysis is based on the well-known framework provided by the core tenets of fairness, accountability, and transparency [27,28]. The main contributions in this paper are as follows: • We highlight the potential benefits of ChatGPT in marketing, such as automating tasks, providing more accurate insights, and improving customer engagement. • We emphasize the ethical and societal implications of using ChatGPT, including its potential to perpetuate bias, replace jobs, and create dependencies. • We provide recommendations for companies to mitigate these potential risks, includ- ing appropriate design and testing, protection of personal data, and responsible use of the technology. 2. Implication of ChatGPT in the Field of Marketing The most popular AI tools the general public uses are voice assistance tools and Smart Home technology such as Siri and Amazon Alexa. There has been a lot of academic research on the usage of AI and its impacts on consumers and other stakeholders [29]. The most recent development of ChatGPT has inevitably generated considerable attention and antici- pation in marketing for both academia and practitioners [30]. According to the American Marketing Association, marketing is “the activity, set of institutions, and processes for cre- ating, communicating, delivering, and exchanging offerings that have value for customers, clients, partners, and society at large. (Approved 2017)” [31]. Major marketing activities include strategic planning, marketing research and analysis, marketing strategy decisions in choosing the target market, developing competitive advantage, and developing and main- taining customer relationships [32]. Marketers, customers, clients, partners, and society operate in an environment where disruptive technologies change how businesses operate, and how people live their lives. Among all disruptive technologies, such as the internet of things, big data analytics, blockchain, and artificial intelligence, artificial intelligence (AI) has the highest potential to transform marketing and shape the future of marketing [33–36]. 2.1. What Does the Most Recent Development of ChatGPT Mean in the Field of Marketing? Using ChatGPT, marketing content such as product descriptions and promotional messages can be created with less time and potentially with higher quality than human content creators. Consumer feedback and social media conversations can be collected, thanks to ChatGPT’s capability to condense and analyze large volumes of data [5]. There- fore, marketing research could be conducted more efficiently to understand consumers’ vocabulary, perceptions, and attitudes toward the products and marketing campaigns. Personalized emails and recommendations can be generated and tailored to each customer with little human effort so that each consumer can be treated as an individual with their unique needs and wants being fulfilled. Automated customer service Chatbots can be trained to provide 24/7 support with a human touch so that the issues of generic responses and impersonal tones that existed in traditional chatbots can be overcome. Call center customer service agents can utilize ChatGPT to understand customer questions, identify relevant information, suggest possible solutions, and respond quicker, reducing the cost associated with human intervention and increasing efficiency and accuracy in problem- solving. The new product development team could gather information about current user behavior trends and develop new product innovations based on real-time data collected from different sources by ChatGPT. The above examples are not a comprehensive list of all applications of ChatGPT in marketing. The next few years would witness a giant leap in the marketing field due to the introduction and proper use of ChatGPT. Admittedly, ChatGPT can provide valuable insights and automate specific tasks, but it may also introduce issues or inaccuracies if not used properly [5]. Therefore, it’s essential to carefully consider the data and algorithms used and the ethical implications of the results. Additionally, it’s crucial to validate the results of ChatGPT-generated insights with other data sources and to be transparent about the limitations and uncertainty of the AI 2023, 4 378 results [37]. The following paragraphs briefly discuss such risks, limitations, and a path to move forward. 2.2. Risks of ChatGPT in Marketing AI marketing tools, specifically ChatGPT, can pose several potential risks for mar- keters, consumers, and other stakeholders, as depicted in Figure 1. The potential risks for marketers include the following. First, AI marketing tools such as ChatGPT can pull information from inaccurate sources and therefore provide incorrect information. In the extreme case that inaccurate information reaches the total amount of data, the results will be false [38]. Second, ChatGPT generates responses based on existing information, which is outdated by nature [39]. A forward-looking and proactive marketer would find limited use of outdated data. Third, ChatGPT responses might be similar or homogeneous for the same prompts from different marketers. Therefore, the identity of the marketer or the brand would be impaired. For a field that values creativity and innovation, the issue facing marketers is how to maintain innovativeness in strategic marketing decision-making in an AI-driven world [40]. Fourth, as AI marketing tools such as ChatGPT become more sophisticated, companies may become too reliant on them, leaving them vulnerable if the tools fail or if the data used to train them is inaccurate [41]. Fifth, AI marketing tools such as ChatGPT can lead to job losses as they automate certain marketing tasks previously done by humans [42]. For instance, copy editors and creative writers would shift from more writing to more editing, as faster and more effective writing can be easily generated by ChatGPT. Data analysts and marketing researchers would find data collected and analyzed quickly and automatically. Customer service can be provided 24/7 by interactive and intelligent AI tools rather than human customer service representatives. Sixth, hardly any AI tools can be truly unbiased [40]. AI marketing tools such as ChatGPT can perpetuate and amplify societal biases if they are not properly designed and tested [43]. Seventh, the use of ChatGPT seems intuitive but a certain level of professionalism and expertise is expected for the marketer ’s professional use. The mere use of ChatGPT as a black box should be exercised with caution. Figure 1. An informed understanding of potential risks and a strategic approach to their mitigation are crucial for ensuring the responsible integration of ChatGPT in marketing practices. The risks for consumers include the following. First, AI marketing tools can collect and analyze large amounts of personal data, leading to privacy violations if the data is not adequately protected [44,45]. ChatGPT can learn and adapt to what the user wants and other psychographic characteristics of the user, which implies the user ’s personal data has been widely collected for better responses around iterations of user prompts. Second, although ChatGPT can detect user emotions and provide a more human feeling than traditional AI tools, it is still not human. In the service sectors or areas where human interactions are valued, customers seek a unique and memorable experience, and they do not expect fully automated and emotionless interactions provided by AI marketing [46]. Lack of empathy and understanding of the consumer can lead to consumer disengagement and dissatisfaction [47]. There are risks associated with the application of ChatGPT in marketing. However, these risks can be mitigated through a good and careful design and deployment of ChatGPT and ongoing monitoring and evaluation, as discussed next. AI 2023, 4 379 2.3. Ethical Use of AI-Based Tools in Marketing ChatGPT is expected to become an essential tool for marketing, providing unprece- dented opportunities for personalized, engaging, and effective marketing strategies and research [48]. However, with great power comes great responsibility, and companies must be aware of the ethical considerations of using ChatGPT in marketing. In this article, we discuss several ways to make ethical use of ChatGPT in marketing, including transparency, bias mitigation, privacy protection, risk assessment, accountability, continuous monitoring, ethical decision-making, human oversight, recruiting data science experts, and developing best practices. Transparency is a crucial component of ethical ChatGPT marketing. Companies should be transparent about the data they collect, how it is used, and the algorithms and decision- making processes that use ChatGPT. This ensures that consumers understand how their data is used and can make informed decisions about their privacy. It also promotes trust between companies and consumers, which is essential for long-term customer relationship management. Ma et al. [49], and Rivas et al. [50] suggest that transparency can be achieved by providing precise and concise information to consumers about data collection, data usage, and ChatGPT algorithms. Bias mitigation is another critical consideration in ethical ChatGPT marketing. Compa- nies should actively work to identify and address any biases in their ChatGPT, including through testing and monitoring. This ensures that content created by ChatGPT is inclusive and fair and does not perpetuate existing biases. Akter et al. [51] suggest that companies use diverse data sources and continuously test their ChatGPT for biases. Prioritizing privacy is essential for ethical ChatGPT marketing. Companies should pri- oritize protecting personal data and ensure that their ChatGPT-based technology complies with relevant privacy laws and regulations. Grewal et al. [44] recommend that companies provide consumers with control over their data, such as the ability to opt out of data collection or delete their data. Conducting a risk assessment is a crucial step in ensuring ethical ChatGPT marketing. Companies should conduct a risk assessment to understand ChatGPT’s potential risks and negative consequences and take steps to mitigate them. De Montjoye et al. [52] suggest that companies should assess the potential impact of their ChatGPT on privacy, security, and social welfare. Accountability is essential in ethical ChatGPT marketing. Companies should be ac- countable for the impact of their ChatGPT and be prepared to take corrective action if necessary. Wirtz et al. [53] suggest that businesses need to ensure that they are only collect- ing and using data necessary for their content creation efforts and are responsible for how that data is stored and shared. Continuous monitoring is critical for ethical ChatGPT marketing. Companies should monitor the performance of ChatGPT systems to ensure they are working as expected and make changes as necessary. This includes monitoring for unintended consequences. Bogina et al. [54] suggest that companies continuously test and monitor their ChatGPT usage for ethical and legal compliance. Encouraging ethical decision-making is essential for ethical ChatGPT marketing. Com- panies should establish a clear ethical framework to guide decision-making and train their team on ethical considerations in ChatGPT and content creation. While there are no ethical frameworks specific to ChatGPT, several ethical frameworks are related to language models. One such framework is transcriptivism [55], which deals with the moral responsibility of linguists. Human oversight is also crucial to ensure that the decisions and content created by ChatGPT systems align with ethical guidelines. Companies should have a human oversight system to review and vet the decisions made or content created automatically by ChatGPT systems [56]. AI 2023, 4 380 Recruiting data science experts is essential for many marketing teams, which lack personnel who can understand the data algorithm, data science, and artificial intelligence, creating challenges when using ChatGPT in marketing data analysis and interpretation [57]. Finally, it is crucial to develop best practices in ChatGPT marketing. Rapid learning and experimentation happen during the infancy stage of using ChatGPT in marketing. A well-tested and examined set of practices should be developed and encouraged for more effective and efficient usage of ChatGPT in marketing [58]. It is important to acknowledge that the items discussed above, and summarized in Table 1, are not all-inclusive and will differ based on the specific application of ChatGPT. Moreover, given the rapid evolution of ChatGPT and marketing, it is essential to remain informed about recent advancements, regulations, and recommended protocols. Table 1. Comprehensive Overview of Ethical Considerations and Stakeholders in AI-Based Marketing: A Summary of Strategies, Primary Stakeholders, and References for Responsible AI Implementation in the Marketing Industry. Ways to Make Ethical Use of AI in Marketing Primary Stakeholders References Ensuring transparency Companies, Consumers [49,50] Addressing bias Companies, Consumers [51] Prioritizing privacy Companies, Consumers [44] Conducting a risk assessment Companies, Consumers [52] Being responsible Companies, Consumers [53] Continuous monitoring Companies [54] Encouraging ethical decision-making Companies [59] Human oversight Companies, Consumers [56] Recruit data science experts Companies [57] Develop best practices AI marketing Companies [58] 2.4. Ethics Evaluation of ChatGPT Several key factors must be considered when evaluating the ethics of an AI tools such as ChatGPT. These include assessing its potential risks, harms, and benefits, as well as taking steps to mitigate any harmful consequences that may arise. To begin with, companies should conduct a risk assessment to understand ChatGPT’s potential risks and adverse effects and take steps to mitigate them [60]. This is an essential first step in the process, as it allows companies to identify and address any potential issues before they become significant problems. In addition to assessing potential risks, it is also crucial to consider the potential harms that ChatGPT may cause and its benefits. This requires a careful examination of the tool’s capabilities, as well as its potential impact on users, both positive and negative. To ensure that all perspectives are considered, it is important to involve a diverse group of stakeholders in developing and evaluating the tool. Once ChatGPT has been developed and implemented, it is crucial to continuously monitor and evaluate its performance and impact, making necessary adjustments along the way. This helps to ensure that any issues that arise can be addressed promptly and that the tool remains effective and ethical over time. When it comes to marketing ChatGPT solutions, transparency is critical. Companies should be transparent about the capabilities and limitations of their ChatGPT tool, as well as its potential risks and harms. To achieve this, there are several ethical considerations that companies should keep in mind. First and foremost, companies should be transparent about the data they use to train their ChatGPT models, the algorithms and decision-making processes used, and the AI 2023, 4 381 potential risks and harms associated with the ChatGPT tool [61]. This helps ensure that users clearly understand how the tool works and what it can and cannot do. In addition, companies should avoid overstating the capabilities of their ChatGPT tool and provide realistic and accurate information about what it can and cannot do. This helps ensure users have realistic expectations about what the tool can accomplish. It is also important for companies to be mindful of their audience when marketing ChatGPT. This means ensuring that the marketing materials and communications are appro- priate for the intended audience and that users understand how to use the tool effectively. Companies should also be honest about their ChatGPT tool’s limitations and potential risks [61]. By acknowledging these issues upfront, companies can help build trust with users and avoid potential ethical issues. Finally, companies should be accountable for the impact of their ChatGPT tool and be prepared to take corrective action if necessary [5]. This means being proactive in identifying and addressing any ethical issues that arise and being transparent with users about how these issues are being addressed. Companies can help ensure that their ChatGPT solutions are practical and ethical by taking these steps. 3. Conclusions ChatGPT has the potential to revolutionize the marketing industry by automating tasks, providing insights more accurately than previous PLMs, and potentially improving customer engagement. However, there are concerns about its use’s ethical and societal implications. ChatGPT’s ability to collect and analyze large amounts of personal data has raised privacy concerns, and its potential to perpetuate bias requires appropriate design and testing of its derivative products. Furthermore, ChatGPT can lead to job losses as it automates specific tasks previously done by humans. Despite these potential risks, ChatGPT has generated significant attention and anticipation in the marketing field for academia and practitioners. AI, especially ChatGPT and GPT4, have the highest potential to transform marketing and shape its future. ChatGPT can help marketers create content faster and potentially with quality as good as human content creators. It can also help marketers conduct more efficient re- search and understand consumer vocabulary, perceptions, and attitudes toward products and campaigns. ChatGPT can also help marketers personalize emails and recommen- dations, provide 24/7 automated customer service, and improve call center customer service efficiency and accuracy. However, it is crucial to prioritize transparency, bias mitigation, privacy protection, risk assessment, accountability, continuous monitoring, and ethical decision-making in ChatGPT-based marketing. AI marketing tools based on ChatGPT can pose several potential risks for marketers, consumers, and other stakeholders, such as inaccuracies in the timeliness of its training data, dependency, job replacement, and societal bias. Therefore, companies must carefully consider the potential risks and benefits of Chat- GPT before implementing it in their marketing strategies. Developing best practices in ChatGPT marketing is crucial, and recruiting data science experts is essential for many marketing teams. By prioritizing ethical considerations and human oversight, ChatGPT can help revolutionize marketing while avoiding potential harm to stakeholders. Human oversight is crucial to ensure that the decisions and content created by ChatGPT systems align with ethical guidelines set forth by stakeholders or governments. In conclusion, while ChatGPT has the potential to transform marketing, its implementation must be carefully considered to avoid potential risks and ensure ethical decision-making. Author Contributions: Conceptualization, P.R.; methodology, P.R. and L.Z.; validation, L.Z., investi- gation, P.R. and L.Z.; writing—original draft preparation, P.R.; writing—review and editing, L.Z. and P.R. All authors have read and agreed to the published version of the manuscript. Funding: This research was executed while P.R. was funded by the National Science Foundation under grant NSF CISE-CNS Award 2136961. AI 2023, 4 382 Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: No new data were created or analyzed in this study. Data sharing is not applicable to this article. Acknowledgments: We want to thank the reviewers for their time and valuable feedback. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the writing of the manuscript; or in the decision to publish the results. References 1. Chen, C. China’s ChatGPT Black Market Is Thriving. Available online: https://www.wired.com/story/chinas-chatgpt-black- market-baidu/ (accessed on 7 March 2023). 2. Kumar, A.H. Analysis of ChatGPT Tool to Assess the Potential of its Utility for Academic Writing in Biomedical Domain. Biol. Eng. Med. Sci. Rep. 2023, 9, 24–30. [CrossRef] 3. Scaringi, G.; Loche, M. An interview with ChatGPT: Discussing artificial intelligence in teaching, research, and practice. arXiv 2023. [CrossRef] 4. Sakirin, T.; Said, R.B. User preferences for ChatGPT-powered conversational interfaces versus traditional methods. Mesopotamian J. Comput. Sci. 2023, 2023, 24–31. [CrossRef] [PubMed] 5. Zielinski, C.; Winker, M.; Aggarwal, R.; Ferris, L.; Heinemann, M.; Lapeña, J.F.; Pai, S.; Citrome, L. Chatbots, ChatGPT, and Scholarly Manuscripts-WAME Recommendations on ChatGPT and Chatbots in Relation to Scholarly Publications. Afro-Egypt. J. Infect. Endem. Dis. 2023, 13, 75–79. 6. Stokel-Walker, C.; Van Noorden, R. What ChatGPT and generative AI mean for science. Nature 2023, 614, 214–216. [CrossRef] [PubMed] 7. Patel, S.B.; Lam, K.; Liebrenz, M. ChatGPT: Friend or Foe. Lancet Digit. Health 2023, 5, e102. 8. Mijwil, M.; Aljanabi, M.; Ali, A.H. ChatGPT: Exploring the Role of Cybersecurity in the Protection of Medical Information. Mesopotamian J. Cybersecur. 2023, 2023, 18–21. [CrossRef] 9. George, A.S.; George, A.H.; Martin, A.G. A Review of ChatGPT AI’s Impact on Several Business Sectors. Partners Univers. Int. Innov. J. 2023, 1, 1. 10. Bornstein, M.; Stahl, S. ChatGPT vs. Content Marketing, A Free Webinar Brought to You by ON . Available online: https: //contentmarketinginstitute.com/ (accessed on 2 February 2023). 11. Bowman, E. A New AI Chatbot Might do Your Homework for you. But It’s Still not an A+ Student. NPR. Available online: https://www.npr.org/2022/12/19/1143912956/chatgpt-ai-chatbot-homework-academia (accessed on 2 January 2023). 12. Thorp, H.H. ChatGPT is Fun, But Not an Author. Science. https://www.science.org/doi/10.1126/science.adg7879 (accessed on 28 January 2023). 13. Zhuo, T.Y.; Huang, Y.; Chen, C.; Xing, Z. Exploring ai ethics of chatgpt: A diagnostic analysis. arXiv 2023, arXiv:2301.12867. 14. Rivas, P.; Holzmayer, K.; Hernandez, C.; Grippaldi, C. Excitement and concerns about machine learning-based chatbots and talkbots: A survey. In Proceedings of the 2018 IEEE International Symposium on Technology and Society (ISTAS), Washington, DC, USA, 13–14 November 2018; pp. 156–162. 15. Plant, R.; Giuffrida, V.; Gkatzia, D. You Are What You Write: Preserving Privacy in the Era of Large Language Models. arXiv 2022, arXiv:2204.09391. 16. Yuan, S.; Zhao, H.; Zhao, S.; Leng, J.; Liang, Y.; Wang, X.; Yu, J.; Lv, X.; Shao, Z.; He, J.; et al. A roadmap for big model. arXiv 2022, arXiv:2203.14101 17. Ahn, J.; Oh, A. Mitigating language-dependent ethnic bias in BERT. arXiv 2021, arXiv:2109.05704. 18. Gao, L.; Zhan, H.; Chen, A.; Sheng, V. Mitigate Gender Bias using Negative Multi-Task Learning. 2022. Preprint [CrossRef] 19. Kirov, V.; Malamin, B. Are Translators Afraid of Artificial Intelligence? Societies 2022, 12, 70. [CrossRef] 20. Abdullah, Y.I.; Schuman, J.S.; Shabsigh, R.; Caplan, A.; Al-Aswad, L.A. Ethics of artificial intelligence in medicine and ophthalmology. Asia-Pac. J. Ophthalmol. 2021, 10, 289. [CrossRef] [PubMed] 21. Hacker, P.; Engel, A.; Mauer, M. Regulating ChatGPT and other Large Generative AI Models. arXiv 2023, arXiv:2302.02337. 22. Selvaraju, R.R.; Lee, S.; Shen, Y.; Jin, H.; Ghosh, S.; Heck, L.; Batra, D.; Parikh, D. Taking a hint: Leveraging explanations to make vision and language models more grounded. In Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Republic of Korea, 27 October–2 November 2019; pp. 2591–2600. 23. Borji, A. A categorical archive of ChatGPT failures. arXiv 2023, arXiv:2302.03494. 24. Dziri, N.; Milton, S.; Yu, M.; Zaiane, O.; Reddy, S. On the origin of hallucinations in conversational models: Is it the datasets or the models? arXiv 2022, arXiv:2204.07931. 25. Kasneci, E.; Seßler, K.; Küchemann, S.; Bannert, M.; Dementieva, D.; Fischer, F.; Gasser, U.; Groh, G.; Günnemann, S.; Hüllermeier, E.; et al. ChatGPT for good? On opportunities and challenges of large language models for education. Learn. Individ. Differ. 2023, 103, 102274. [CrossRef] AI 2023, 4 383 26. Schramowski, P.; Turan, C.; Andersen, N.; Rothkopf, C.A.; Kersting, K. Large pre-trained language models contain human-like biases of what is right and wrong to do. Nat. Mach. Intell. 2022, 4, 258–268. [CrossRef] 27. Zhou, J.; Chen, F.; Berry, A.; Reed, M.; Zhang, S.; Savage, S. A survey on ethical principles of AI and implementations. In Proceedings of the 2020 IEEE Symposium Series on Computational Intelligence (SSCI), Canberra, Australia, 1–4 December 2020; pp. 3010–3017. 28. Häußermann, J.J.; Lütge, C. Community-in-the-loop: Towards pluralistic value creation in AI, or—why AI needs business ethics. Ethics 2022, 2022, 1–22. [CrossRef] 29. Hosseini, M.; Horbach, S.P. Fighting reviewer fatigue or amplifying bias? Considerations and recommendations for use of ChatGPT and other Large Language Models in scholarly peer review. Res. Sq. 2023. Preprint. 30. Azaria, A. ChatGPT Usage and Limitations 2022. Preprint 2022. Available online: https://hal.science/hal-03913837/ (accessed on 2 February 2023). 31. Association, A.M. Definition of Marketing. Available online: https://www.ama.org/the-definition-of-marketing-what-is- marketing/ (accessed on 20 February 2023). 32. Ferrell, O.C.; Hartline, M.; Hochstein, B.W. Marketing Strategy, 8th ed.; Cengage Learning: Boston, MS, USA, 2021. 33. Verma, S.; Sharma, R.; Deb, S.; Maitra, D. Artificial intelligence in marketing: Systematic review and future research direction. Int. J. Inf. Manag. Data Insights 2021, 1, 100002. [CrossRef] 34. Oberoi, P. How Artificial Intelligence Is Impacting Marketing? In Encyclopedia of Data Science and Machine Learning; Wang, J., ed.; IGI Global: Hershey, PA, USA, 2023; pp. 606–618. [CrossRef] 35. Pitt, C.; Paschen, J.; Kietzmann, J.; Pitt, L.F.; Pala, E. Artificial intelligence, marketing, and the history of technology: Kranzberg’s laws as a conceptual lens. Australas. Mark. J. 2023, 31, 81–89. [CrossRef] 36. Zaman, K. Transformation of Marketing Decisions through Artificial Intelligence and Digital Marketing. J. Mark. Strateg. 2022, 4, 353–364. [CrossRef] 37. Lahoti, P.; Gummadi, K.P.; Weikum, G. Detecting and mitigating test-time failure risks via model-agnostic uncertainty learning. In Proceedings of the 2021 IEEE International Conference on Data Mining (ICDM), Auckland, New Zealand, 7–10 December 2021; pp. 1174–1179. 38. Yara, O.; Brazheyev, A.; Golovko, L.; Bashkatova, V. Legal Regulation of the Use of Artificial Intelligence: Problems and Development Prospects. Eur. J. Sustain. Dev. 2023, 10, 1. [CrossRef] 39. Sun, G.H.; Hoelscher, S.H. The ChatGPT Storm and What Faculty Can Do. Nurse Educ. 2023. Preprint. [CrossRef] 40. Stone, M.; Aravopoulou, E.; Ekinci, Y.; Evans, G.; Hobbs, M.; Labib, A.; Laughlin, P.; Machtynger, J.; Machtynger, L. Artificial Intelligence (AI) in Strategic Marketing Decision-Making: A research agenda. Bottom Line 2020, 33, 183–200. [CrossRef] 41. Krafft, M.; Sajtos, L.; Haenlein, M. Challenges and opportunities for marketing scholars in times of the fourth industrial revolution. J. Interact. Mark. 2020, 51, 1–8. [CrossRef] 42. Vlaci ˇ c, ´ B.; Corbo, L.; e Silva, S.C.; Dabic, ´ M. The evolving role of artificial intelligence in marketing: A review and research agenda. J. Bus. Res. 2021, 128, 187–203. [CrossRef] 43. Hermann, E. Leveraging artificial intelligence in marketing for social good—An ethical perspective. J. Bus. Ethics 2022, 179, 43–61. [CrossRef] 44. Grewal, D.; Hulland, J.; Kopalle, P.K.; Karahanna, E. The future of technology and marketing: A multidisciplinary perspective. J. Acad. Mark. Sci. 2020, 48, 1–8. [CrossRef] 45. Mogaji, E.; Soetan, T.O.; Kieu, T.A. The implications of artificial intelligence on the digital marketing of financial services to vulnerable customers. Australas. Mark. J. 2020, 29, 235–242. [CrossRef] 46. Paluch, S.; Wirtz, J. Editorial: Artificial Intelligence and Robots in the Service Encounter. J. Serv. Manag. Res. 2020, 4, 3–8. 47. Chowdhury, R.M.; Fernando, M. The Relationships of Empathy, Moral Identity and Cynicism with Consumers’ Ethical Beliefs: The Mediating Role of Moral Disengagement. J. Bus. Ethics 2014, 124, 677–694. [CrossRef] 48. Brand, J.; Israeli, A.; Ngwe, D. Using GPT for Market Research. SSRN. Preprint. 2023. [CrossRef] 49. Ma, L.; Sun, B. Machine learning and AI in marketing–Connecting computing power to human insights. Int. J. Res. Mark. 2020, 37, 481–504. [CrossRef] 50. Rivas, P.; Chelsi, C.; Nishit, N.; Ravula, L. Application-agnostic chatbot deployment considerations: A case study. In Proceedings of the 2019 International Conference on Computational Science and Computational Intelligence (CSCI), Las Vegas, NV, USA, 5–7 December 2019; pp. 361–365. 51. Akter, S.; McCarthy, G.; Sajib, S.; Michael, K.; Dwivedi, Y.K.; D’Ambra, J.; Shen, K.N. Algorithmic bias in data-driven innovation in the age of AI. Int. J. Inf. Manag. 2021, 60, 102387. [CrossRef] 52. De Bruyn, A.; Viswanathan, V.; Beh, Y.S.; Brock, J.K.U.; Von Wangenheim, F. Artificial intelligence and marketing: Pitfalls and opportunities. J. Interact. Mark. 2020, 51, 91–105. [CrossRef] 53. Wirtz, B.W.; Weyerer, J.C.; Sturm, B.J. The dark sides of artificial intelligence: An integrated AI governance framework for public administration. Int. J. Public Adm. 2020, 43, 818–829. [CrossRef] 54. Bogina, V.; Hartman, A.; Kuflik, T.; Shulner-Tal, A. Educating software and AI stakeholders about algorithmic fairness, accountability, transparency and ethics. Int. J. Artif. Intell. Educ. 2021, 32, 808–833 [CrossRef] 55. Kibbey, T. Transcriptivism: An ethical framework for modern linguistics. Proc. Linguist. Soc. Am. 2019, 4, 45. [CrossRef] 56. Etzioni, A.; Etzioni, O. Incorporating ethics into artificial intelligence. J. Ethics 2017, 21, 403–418. [CrossRef] AI 2023, 4 384 57. Agarwal, N. Artificial Intelligence and Marketing. Int. J. Soc. Sci. Econ. Res. 2022, 7. 58. Davenport, T.; Guha, A.; Grewal, D.; Bressgott, T. How artificial intelligence will change the future of marketing. J. Acad. Mark. Sci. 2020, 48, 24–42. [CrossRef] 59. Rivas, P. AI Orthopraxy: Towards a Framework for AI That Promotes Fairness. In Proceedings of the 2020 IEEE International Symposium on Technology and Society (ISTAS), Virtual, 12–15 November 2020; pp. 80–84. 60. Quinn, S.; Souppaya, M.; Cook, M.; Scarfone, K. National Checklist Program for IT Products: Guidelines for Checklist Users and Developers; Technical Report; National Institute of Standards and Technology: Gaithersburg, MD, USA, 2017. [CrossRef] 61. Sallam, M. The Utility of ChatGPT as an Example of Large Language Models in Healthcare Education, Research and Practice: Systematic Review on the Future Perspectives and Potential Limitations. medRxiv 2023. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png AI Multidisciplinary Digital Publishing Institute

Marketing with ChatGPT: Navigating the Ethical Terrain of GPT-Based Chatbot Technology

Rivas, Pablo; Zhao, Liang
AI , Volume 4 (2) – Apr 10, 2023
Download PDF
10 pages

Loading next page...
 
/lp/multidisciplinary-digital-publishing-institute/marketing-with-chatgpt-navigating-the-ethical-terrain-of-gpt-based-8O3m0p6xlV
Publisher
Multidisciplinary Digital Publishing Institute
Copyright
© 1996-2023 MDPI (Basel, Switzerland) unless otherwise stated Disclaimer Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. Terms and Conditions Privacy Policy
ISSN
2673-2688
DOI
10.3390/ai4020019
Publisher site
See Article on Publisher Site

Abstract

Commentary Marketing with ChatGPT: Navigating the Ethical Terrain of GPT-Based Chatbot Technology 1, ,†,‡, 2,‡ Pablo Rivas * and Liang Zhao Department of Computer Science, Baylor University, Waco, TX 76798, USA Department of Marketing, St. Ambrose University, Davenport, IA 52803, USA * Correspondence: pablo_rivas@baylor.edu; Tel.: +1-254-710-3385 † Current address: Center for Standards and Ethics in Artificial Intelligence, Waco, TX 76798, USA. ‡ These authors contributed equally to this work. Abstract: ChatGPT is an AI-powered chatbot platform that enables human users to converse with machines. It utilizes natural language processing and machine learning algorithms, transforming how people interact with AI technology. ChatGPT offers significant advantages over previous similar tools, and its potential for application in various fields has generated attention and anticipation. However, some experts are wary of ChatGPT, citing ethical implications. Therefore, this paper shows that ChatGPT has significant potential to transform marketing and shape its future if certain ethical considerations are taken into account. First, we argue that ChatGPT-based tools can help marketers create content faster and potentially with quality similar to human content creators. It can also assist marketers in conducting more efficient research and understanding customers better, automating customer service, and improving efficiency. Then we discuss ethical implications and potential risks for marketers, consumers, and other stakeholders, that are essential for ChatGPT-based marketing; doing so can help revolutionize marketing while avoiding potential harm to stakeholders. Keywords: marketing; ChatGPT; GPT technology; Chatbot; ethics; societal implications 1. Introduction to ChatGPT Citation: Rivas, P.; Zhao, L. On 30 November 2022, OpenAI unveiled a chatbot named ChatGPT, quickly draw- Marketing with ChatGPT: ing the attention of artificial intelligence (AI) researchers and scholars who tested the Navigating the Ethical Terrain of technology over several hours and days to follow. The release of ChatGPT also sparked GPT-Based Chatbot Technology. AI widespread public interest, with individuals across the globe eager to experience the in- 2023, 4, 375–384. https://doi.org/ novation and assess its capabilities. In regions where ChatGPT is not yet available, such 10.3390/ai4020019 as China, the acquisition of a ChatGPT login has become a highly coveted commodity, fueling the growth of an illicit market in which more than 600 vendors peddle logins Academic Editor: Rafał Drezewski ˙ for a nominal fee [1]. Designed to generate detailed textual responses and articulate an- Received: 18 February 2023 swers to any inquiry [2], ChatGPT has garnered attention for its potential application in Revised: 1 April 2023 the field of engineering geology [3]. One study demonstrated that 70% of users favored Accepted: 4 April 2023 ChatGPT-powered conversational interfaces over conventional methods, citing conve- Published: 10 April 2023 nience, efficiency, and personalization [4]. However, some experts are wary of the tool, with a large AI conference banning its use in conference papers [5]. Additionally, the journal Science has stated that no text generated by ChatGPT or any other AI tool can be used in a paper [6]; Elsevier has established a policy on AI-assisted technologies in scientific Copyright: © 2023 by the authors. writing that requires authors to disclose their use, and not list or cite these tools as authors Licensee MDPI, Basel, Switzerland. or co-authors, limiting the use of AI for enhancing the readability and language of the This article is an open access article work [7]. distributed under the terms and ChatGPT is an AI-powered chatbot platform that utilizes natural language process- conditions of the Creative Commons ing and machine learning algorithms to enable human users to converse with machines, Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ thereby transforming how people interact with AI technology [8]. Before this advancement, 4.0/). most commercialized AI tools were conventional bots or voice assistants with limited AI 2023, 4, 375–384. https://doi.org/10.3390/ai4020019 https://www.mdpi.com/journal/ai AI 2023, 4 376 applications. In contrast, ChatGPT offers numerous advantages over previous AI tools and holds significant potential for application in various fields, including customer service, edu- cation, health care, finance, entertainment, creative writing, digital marketing, e-commerce, and many others [9]. For example, e-commerce companies could leverage ChatGPT to provide automated customer service and order fulfillment services round-the-clock, re- ducing personnel resource costs. Similarly, educators could harness ChatGPT to develop individualized study plans tailored to each student’s interests and progress and to deliver instantaneous feedback through automated grading and virtual assistance in answering questions. In the healthcare sector, healthcare professionals could use ChatGPT to triage symptoms, conduct initial consultations, and even diagnose conditions. The customized and empathetic nature of these conversations may promote a more patient-centered ap- proach and alleviate the anxiety and social pressure that patients may experience when discussing certain health conditions, such as mental illness, with an actual human being [8]. A content marketing company recently conducted an experiment utilizing ChatGPT to create promotional Twitter posts advertising an upcoming webinar on whether robots will eventually replace human labor in content marketing [10]. Here are examples of the two generated Twitter posts: Can you tell the difference between human-written and AI-generated content? Put your skills to the test in our upcoming webinar “ChatGPT vs. Content Marketing” on 15 Feb at 2 p.m. EST. Register now! #AI #conTentmarketing #webinar. The future of marketing is here and it’s all about AI. Don’t miss our webinar “Chat GPT vs. Content Marketing” on Wed. 15 Feb at 2 p.m. EST to learn how to harness the power of AI tools like ChatGPT for your content marketing. Register now! #AI #contentmarketing #webinar. The AI-generated posts were completed much more efficiently than human-generated posts, and even suggested hashtags for users. The list of potential benefits goes on [11]. While these benefits are promising, the tool’s limitations and potential ethical implications must also be considered. Despite the excitement surrounding ChatGPT and its numerous applications, some individuals have expressed ethical concerns regarding this latest inno- vation [12,13]. More specifically, there are known risks associated with AI-based chatbots; for ChatGPT, these mainly include the following [11,14]. ChatGPT can collect and analyze large amounts of personal data, leading to privacy violations if the data is not adequately protected [15,16]. Bias can also be perpetuated and amplified by ChatGPT if it is not ad- equately designed and tested [17,18]. In addition to these risks, ChatGPT can lead to job losses as it automates specific tasks previously done by humans [19]. As ChatGPT becomes more sophisticated, companies may become too reliant on it, leaving them vulnerable if the tool fails or if the data used to train it is inaccurate [20]. Furthermore, the tool has been perceived as a new way to cheat, making the AI write the work and submit it as your own for credit, which generally constitutes plagiarism [12]. Fur- thermore, training ChatGPT has caused severe mental stress on its content curators [21,22]. ChatGPT can generate incorrect, incomplete, misleading, biased, harmful information, instructions, or content [23]. In addition, it sometimes generates fake outputs that are not real, which can erode the trust in citing science [24]. Finally, we emphasize that ChatGPT does not have a human oversight system to review and vet all its outputs, which can lead to ethical violations [25]. It is also important to note that ChatGPT is not unique in its potential for ethical implications. Instead, it is a representative example of large-scale pre-trained language models (PLMs), a rapidly developing field in Natural Language Processing (NLP) [26]. To illustrate the potential ethical implications of large-scale PLMs, we chose to focus on ChatGPT as a case study, examining its specific characteristics and design to draw broader implications for large-scale PLMs, although our focus is on marketing. In the following section, we will explore in detail the implications of ChatGPT in the field of marketing and how it can be used to address some of the ethical concerns while maximizing its benefits. AI 2023, 4 377 This ethical analysis is based on the well-known framework provided by the core tenets of fairness, accountability, and transparency [27,28]. The main contributions in this paper are as follows: • We highlight the potential benefits of ChatGPT in marketing, such as automating tasks, providing more accurate insights, and improving customer engagement. • We emphasize the ethical and societal implications of using ChatGPT, including its potential to perpetuate bias, replace jobs, and create dependencies. • We provide recommendations for companies to mitigate these potential risks, includ- ing appropriate design and testing, protection of personal data, and responsible use of the technology. 2. Implication of ChatGPT in the Field of Marketing The most popular AI tools the general public uses are voice assistance tools and Smart Home technology such as Siri and Amazon Alexa. There has been a lot of academic research on the usage of AI and its impacts on consumers and other stakeholders [29]. The most recent development of ChatGPT has inevitably generated considerable attention and antici- pation in marketing for both academia and practitioners [30]. According to the American Marketing Association, marketing is “the activity, set of institutions, and processes for cre- ating, communicating, delivering, and exchanging offerings that have value for customers, clients, partners, and society at large. (Approved 2017)” [31]. Major marketing activities include strategic planning, marketing research and analysis, marketing strategy decisions in choosing the target market, developing competitive advantage, and developing and main- taining customer relationships [32]. Marketers, customers, clients, partners, and society operate in an environment where disruptive technologies change how businesses operate, and how people live their lives. Among all disruptive technologies, such as the internet of things, big data analytics, blockchain, and artificial intelligence, artificial intelligence (AI) has the highest potential to transform marketing and shape the future of marketing [33–36]. 2.1. What Does the Most Recent Development of ChatGPT Mean in the Field of Marketing? Using ChatGPT, marketing content such as product descriptions and promotional messages can be created with less time and potentially with higher quality than human content creators. Consumer feedback and social media conversations can be collected, thanks to ChatGPT’s capability to condense and analyze large volumes of data [5]. There- fore, marketing research could be conducted more efficiently to understand consumers’ vocabulary, perceptions, and attitudes toward the products and marketing campaigns. Personalized emails and recommendations can be generated and tailored to each customer with little human effort so that each consumer can be treated as an individual with their unique needs and wants being fulfilled. Automated customer service Chatbots can be trained to provide 24/7 support with a human touch so that the issues of generic responses and impersonal tones that existed in traditional chatbots can be overcome. Call center customer service agents can utilize ChatGPT to understand customer questions, identify relevant information, suggest possible solutions, and respond quicker, reducing the cost associated with human intervention and increasing efficiency and accuracy in problem- solving. The new product development team could gather information about current user behavior trends and develop new product innovations based on real-time data collected from different sources by ChatGPT. The above examples are not a comprehensive list of all applications of ChatGPT in marketing. The next few years would witness a giant leap in the marketing field due to the introduction and proper use of ChatGPT. Admittedly, ChatGPT can provide valuable insights and automate specific tasks, but it may also introduce issues or inaccuracies if not used properly [5]. Therefore, it’s essential to carefully consider the data and algorithms used and the ethical implications of the results. Additionally, it’s crucial to validate the results of ChatGPT-generated insights with other data sources and to be transparent about the limitations and uncertainty of the AI 2023, 4 378 results [37]. The following paragraphs briefly discuss such risks, limitations, and a path to move forward. 2.2. Risks of ChatGPT in Marketing AI marketing tools, specifically ChatGPT, can pose several potential risks for mar- keters, consumers, and other stakeholders, as depicted in Figure 1. The potential risks for marketers include the following. First, AI marketing tools such as ChatGPT can pull information from inaccurate sources and therefore provide incorrect information. In the extreme case that inaccurate information reaches the total amount of data, the results will be false [38]. Second, ChatGPT generates responses based on existing information, which is outdated by nature [39]. A forward-looking and proactive marketer would find limited use of outdated data. Third, ChatGPT responses might be similar or homogeneous for the same prompts from different marketers. Therefore, the identity of the marketer or the brand would be impaired. For a field that values creativity and innovation, the issue facing marketers is how to maintain innovativeness in strategic marketing decision-making in an AI-driven world [40]. Fourth, as AI marketing tools such as ChatGPT become more sophisticated, companies may become too reliant on them, leaving them vulnerable if the tools fail or if the data used to train them is inaccurate [41]. Fifth, AI marketing tools such as ChatGPT can lead to job losses as they automate certain marketing tasks previously done by humans [42]. For instance, copy editors and creative writers would shift from more writing to more editing, as faster and more effective writing can be easily generated by ChatGPT. Data analysts and marketing researchers would find data collected and analyzed quickly and automatically. Customer service can be provided 24/7 by interactive and intelligent AI tools rather than human customer service representatives. Sixth, hardly any AI tools can be truly unbiased [40]. AI marketing tools such as ChatGPT can perpetuate and amplify societal biases if they are not properly designed and tested [43]. Seventh, the use of ChatGPT seems intuitive but a certain level of professionalism and expertise is expected for the marketer ’s professional use. The mere use of ChatGPT as a black box should be exercised with caution. Figure 1. An informed understanding of potential risks and a strategic approach to their mitigation are crucial for ensuring the responsible integration of ChatGPT in marketing practices. The risks for consumers include the following. First, AI marketing tools can collect and analyze large amounts of personal data, leading to privacy violations if the data is not adequately protected [44,45]. ChatGPT can learn and adapt to what the user wants and other psychographic characteristics of the user, which implies the user ’s personal data has been widely collected for better responses around iterations of user prompts. Second, although ChatGPT can detect user emotions and provide a more human feeling than traditional AI tools, it is still not human. In the service sectors or areas where human interactions are valued, customers seek a unique and memorable experience, and they do not expect fully automated and emotionless interactions provided by AI marketing [46]. Lack of empathy and understanding of the consumer can lead to consumer disengagement and dissatisfaction [47]. There are risks associated with the application of ChatGPT in marketing. However, these risks can be mitigated through a good and careful design and deployment of ChatGPT and ongoing monitoring and evaluation, as discussed next. AI 2023, 4 379 2.3. Ethical Use of AI-Based Tools in Marketing ChatGPT is expected to become an essential tool for marketing, providing unprece- dented opportunities for personalized, engaging, and effective marketing strategies and research [48]. However, with great power comes great responsibility, and companies must be aware of the ethical considerations of using ChatGPT in marketing. In this article, we discuss several ways to make ethical use of ChatGPT in marketing, including transparency, bias mitigation, privacy protection, risk assessment, accountability, continuous monitoring, ethical decision-making, human oversight, recruiting data science experts, and developing best practices. Transparency is a crucial component of ethical ChatGPT marketing. Companies should be transparent about the data they collect, how it is used, and the algorithms and decision- making processes that use ChatGPT. This ensures that consumers understand how their data is used and can make informed decisions about their privacy. It also promotes trust between companies and consumers, which is essential for long-term customer relationship management. Ma et al. [49], and Rivas et al. [50] suggest that transparency can be achieved by providing precise and concise information to consumers about data collection, data usage, and ChatGPT algorithms. Bias mitigation is another critical consideration in ethical ChatGPT marketing. Compa- nies should actively work to identify and address any biases in their ChatGPT, including through testing and monitoring. This ensures that content created by ChatGPT is inclusive and fair and does not perpetuate existing biases. Akter et al. [51] suggest that companies use diverse data sources and continuously test their ChatGPT for biases. Prioritizing privacy is essential for ethical ChatGPT marketing. Companies should pri- oritize protecting personal data and ensure that their ChatGPT-based technology complies with relevant privacy laws and regulations. Grewal et al. [44] recommend that companies provide consumers with control over their data, such as the ability to opt out of data collection or delete their data. Conducting a risk assessment is a crucial step in ensuring ethical ChatGPT marketing. Companies should conduct a risk assessment to understand ChatGPT’s potential risks and negative consequences and take steps to mitigate them. De Montjoye et al. [52] suggest that companies should assess the potential impact of their ChatGPT on privacy, security, and social welfare. Accountability is essential in ethical ChatGPT marketing. Companies should be ac- countable for the impact of their ChatGPT and be prepared to take corrective action if necessary. Wirtz et al. [53] suggest that businesses need to ensure that they are only collect- ing and using data necessary for their content creation efforts and are responsible for how that data is stored and shared. Continuous monitoring is critical for ethical ChatGPT marketing. Companies should monitor the performance of ChatGPT systems to ensure they are working as expected and make changes as necessary. This includes monitoring for unintended consequences. Bogina et al. [54] suggest that companies continuously test and monitor their ChatGPT usage for ethical and legal compliance. Encouraging ethical decision-making is essential for ethical ChatGPT marketing. Com- panies should establish a clear ethical framework to guide decision-making and train their team on ethical considerations in ChatGPT and content creation. While there are no ethical frameworks specific to ChatGPT, several ethical frameworks are related to language models. One such framework is transcriptivism [55], which deals with the moral responsibility of linguists. Human oversight is also crucial to ensure that the decisions and content created by ChatGPT systems align with ethical guidelines. Companies should have a human oversight system to review and vet the decisions made or content created automatically by ChatGPT systems [56]. AI 2023, 4 380 Recruiting data science experts is essential for many marketing teams, which lack personnel who can understand the data algorithm, data science, and artificial intelligence, creating challenges when using ChatGPT in marketing data analysis and interpretation [57]. Finally, it is crucial to develop best practices in ChatGPT marketing. Rapid learning and experimentation happen during the infancy stage of using ChatGPT in marketing. A well-tested and examined set of practices should be developed and encouraged for more effective and efficient usage of ChatGPT in marketing [58]. It is important to acknowledge that the items discussed above, and summarized in Table 1, are not all-inclusive and will differ based on the specific application of ChatGPT. Moreover, given the rapid evolution of ChatGPT and marketing, it is essential to remain informed about recent advancements, regulations, and recommended protocols. Table 1. Comprehensive Overview of Ethical Considerations and Stakeholders in AI-Based Marketing: A Summary of Strategies, Primary Stakeholders, and References for Responsible AI Implementation in the Marketing Industry. Ways to Make Ethical Use of AI in Marketing Primary Stakeholders References Ensuring transparency Companies, Consumers [49,50] Addressing bias Companies, Consumers [51] Prioritizing privacy Companies, Consumers [44] Conducting a risk assessment Companies, Consumers [52] Being responsible Companies, Consumers [53] Continuous monitoring Companies [54] Encouraging ethical decision-making Companies [59] Human oversight Companies, Consumers [56] Recruit data science experts Companies [57] Develop best practices AI marketing Companies [58] 2.4. Ethics Evaluation of ChatGPT Several key factors must be considered when evaluating the ethics of an AI tools such as ChatGPT. These include assessing its potential risks, harms, and benefits, as well as taking steps to mitigate any harmful consequences that may arise. To begin with, companies should conduct a risk assessment to understand ChatGPT’s potential risks and adverse effects and take steps to mitigate them [60]. This is an essential first step in the process, as it allows companies to identify and address any potential issues before they become significant problems. In addition to assessing potential risks, it is also crucial to consider the potential harms that ChatGPT may cause and its benefits. This requires a careful examination of the tool’s capabilities, as well as its potential impact on users, both positive and negative. To ensure that all perspectives are considered, it is important to involve a diverse group of stakeholders in developing and evaluating the tool. Once ChatGPT has been developed and implemented, it is crucial to continuously monitor and evaluate its performance and impact, making necessary adjustments along the way. This helps to ensure that any issues that arise can be addressed promptly and that the tool remains effective and ethical over time. When it comes to marketing ChatGPT solutions, transparency is critical. Companies should be transparent about the capabilities and limitations of their ChatGPT tool, as well as its potential risks and harms. To achieve this, there are several ethical considerations that companies should keep in mind. First and foremost, companies should be transparent about the data they use to train their ChatGPT models, the algorithms and decision-making processes used, and the AI 2023, 4 381 potential risks and harms associated with the ChatGPT tool [61]. This helps ensure that users clearly understand how the tool works and what it can and cannot do. In addition, companies should avoid overstating the capabilities of their ChatGPT tool and provide realistic and accurate information about what it can and cannot do. This helps ensure users have realistic expectations about what the tool can accomplish. It is also important for companies to be mindful of their audience when marketing ChatGPT. This means ensuring that the marketing materials and communications are appro- priate for the intended audience and that users understand how to use the tool effectively. Companies should also be honest about their ChatGPT tool’s limitations and potential risks [61]. By acknowledging these issues upfront, companies can help build trust with users and avoid potential ethical issues. Finally, companies should be accountable for the impact of their ChatGPT tool and be prepared to take corrective action if necessary [5]. This means being proactive in identifying and addressing any ethical issues that arise and being transparent with users about how these issues are being addressed. Companies can help ensure that their ChatGPT solutions are practical and ethical by taking these steps. 3. Conclusions ChatGPT has the potential to revolutionize the marketing industry by automating tasks, providing insights more accurately than previous PLMs, and potentially improving customer engagement. However, there are concerns about its use’s ethical and societal implications. ChatGPT’s ability to collect and analyze large amounts of personal data has raised privacy concerns, and its potential to perpetuate bias requires appropriate design and testing of its derivative products. Furthermore, ChatGPT can lead to job losses as it automates specific tasks previously done by humans. Despite these potential risks, ChatGPT has generated significant attention and anticipation in the marketing field for academia and practitioners. AI, especially ChatGPT and GPT4, have the highest potential to transform marketing and shape its future. ChatGPT can help marketers create content faster and potentially with quality as good as human content creators. It can also help marketers conduct more efficient re- search and understand consumer vocabulary, perceptions, and attitudes toward products and campaigns. ChatGPT can also help marketers personalize emails and recommen- dations, provide 24/7 automated customer service, and improve call center customer service efficiency and accuracy. However, it is crucial to prioritize transparency, bias mitigation, privacy protection, risk assessment, accountability, continuous monitoring, and ethical decision-making in ChatGPT-based marketing. AI marketing tools based on ChatGPT can pose several potential risks for marketers, consumers, and other stakeholders, such as inaccuracies in the timeliness of its training data, dependency, job replacement, and societal bias. Therefore, companies must carefully consider the potential risks and benefits of Chat- GPT before implementing it in their marketing strategies. Developing best practices in ChatGPT marketing is crucial, and recruiting data science experts is essential for many marketing teams. By prioritizing ethical considerations and human oversight, ChatGPT can help revolutionize marketing while avoiding potential harm to stakeholders. Human oversight is crucial to ensure that the decisions and content created by ChatGPT systems align with ethical guidelines set forth by stakeholders or governments. In conclusion, while ChatGPT has the potential to transform marketing, its implementation must be carefully considered to avoid potential risks and ensure ethical decision-making. Author Contributions: Conceptualization, P.R.; methodology, P.R. and L.Z.; validation, L.Z., investi- gation, P.R. and L.Z.; writing—original draft preparation, P.R.; writing—review and editing, L.Z. and P.R. All authors have read and agreed to the published version of the manuscript. Funding: This research was executed while P.R. was funded by the National Science Foundation under grant NSF CISE-CNS Award 2136961. AI 2023, 4 382 Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: No new data were created or analyzed in this study. Data sharing is not applicable to this article. Acknowledgments: We want to thank the reviewers for their time and valuable feedback. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the writing of the manuscript; or in the decision to publish the results. References 1. Chen, C. China’s ChatGPT Black Market Is Thriving. Available online: https://www.wired.com/story/chinas-chatgpt-black- market-baidu/ (accessed on 7 March 2023). 2. Kumar, A.H. Analysis of ChatGPT Tool to Assess the Potential of its Utility for Academic Writing in Biomedical Domain. Biol. Eng. Med. Sci. Rep. 2023, 9, 24–30. [CrossRef] 3. Scaringi, G.; Loche, M. An interview with ChatGPT: Discussing artificial intelligence in teaching, research, and practice. arXiv 2023. [CrossRef] 4. Sakirin, T.; Said, R.B. User preferences for ChatGPT-powered conversational interfaces versus traditional methods. Mesopotamian J. Comput. Sci. 2023, 2023, 24–31. [CrossRef] [PubMed] 5. Zielinski, C.; Winker, M.; Aggarwal, R.; Ferris, L.; Heinemann, M.; Lapeña, J.F.; Pai, S.; Citrome, L. Chatbots, ChatGPT, and Scholarly Manuscripts-WAME Recommendations on ChatGPT and Chatbots in Relation to Scholarly Publications. Afro-Egypt. J. Infect. Endem. Dis. 2023, 13, 75–79. 6. Stokel-Walker, C.; Van Noorden, R. What ChatGPT and generative AI mean for science. Nature 2023, 614, 214–216. [CrossRef] [PubMed] 7. Patel, S.B.; Lam, K.; Liebrenz, M. ChatGPT: Friend or Foe. Lancet Digit. Health 2023, 5, e102. 8. Mijwil, M.; Aljanabi, M.; Ali, A.H. ChatGPT: Exploring the Role of Cybersecurity in the Protection of Medical Information. Mesopotamian J. Cybersecur. 2023, 2023, 18–21. [CrossRef] 9. George, A.S.; George, A.H.; Martin, A.G. A Review of ChatGPT AI’s Impact on Several Business Sectors. Partners Univers. Int. Innov. J. 2023, 1, 1. 10. Bornstein, M.; Stahl, S. ChatGPT vs. Content Marketing, A Free Webinar Brought to You by ON . Available online: https: //contentmarketinginstitute.com/ (accessed on 2 February 2023). 11. Bowman, E. A New AI Chatbot Might do Your Homework for you. But It’s Still not an A+ Student. NPR. Available online: https://www.npr.org/2022/12/19/1143912956/chatgpt-ai-chatbot-homework-academia (accessed on 2 January 2023). 12. Thorp, H.H. ChatGPT is Fun, But Not an Author. Science. https://www.science.org/doi/10.1126/science.adg7879 (accessed on 28 January 2023). 13. Zhuo, T.Y.; Huang, Y.; Chen, C.; Xing, Z. Exploring ai ethics of chatgpt: A diagnostic analysis. arXiv 2023, arXiv:2301.12867. 14. Rivas, P.; Holzmayer, K.; Hernandez, C.; Grippaldi, C. Excitement and concerns about machine learning-based chatbots and talkbots: A survey. In Proceedings of the 2018 IEEE International Symposium on Technology and Society (ISTAS), Washington, DC, USA, 13–14 November 2018; pp. 156–162. 15. Plant, R.; Giuffrida, V.; Gkatzia, D. You Are What You Write: Preserving Privacy in the Era of Large Language Models. arXiv 2022, arXiv:2204.09391. 16. Yuan, S.; Zhao, H.; Zhao, S.; Leng, J.; Liang, Y.; Wang, X.; Yu, J.; Lv, X.; Shao, Z.; He, J.; et al. A roadmap for big model. arXiv 2022, arXiv:2203.14101 17. Ahn, J.; Oh, A. Mitigating language-dependent ethnic bias in BERT. arXiv 2021, arXiv:2109.05704. 18. Gao, L.; Zhan, H.; Chen, A.; Sheng, V. Mitigate Gender Bias using Negative Multi-Task Learning. 2022. Preprint [CrossRef] 19. Kirov, V.; Malamin, B. Are Translators Afraid of Artificial Intelligence? Societies 2022, 12, 70. [CrossRef] 20. Abdullah, Y.I.; Schuman, J.S.; Shabsigh, R.; Caplan, A.; Al-Aswad, L.A. Ethics of artificial intelligence in medicine and ophthalmology. Asia-Pac. J. Ophthalmol. 2021, 10, 289. [CrossRef] [PubMed] 21. Hacker, P.; Engel, A.; Mauer, M. Regulating ChatGPT and other Large Generative AI Models. arXiv 2023, arXiv:2302.02337. 22. Selvaraju, R.R.; Lee, S.; Shen, Y.; Jin, H.; Ghosh, S.; Heck, L.; Batra, D.; Parikh, D. Taking a hint: Leveraging explanations to make vision and language models more grounded. In Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Republic of Korea, 27 October–2 November 2019; pp. 2591–2600. 23. Borji, A. A categorical archive of ChatGPT failures. arXiv 2023, arXiv:2302.03494. 24. Dziri, N.; Milton, S.; Yu, M.; Zaiane, O.; Reddy, S. On the origin of hallucinations in conversational models: Is it the datasets or the models? arXiv 2022, arXiv:2204.07931. 25. Kasneci, E.; Seßler, K.; Küchemann, S.; Bannert, M.; Dementieva, D.; Fischer, F.; Gasser, U.; Groh, G.; Günnemann, S.; Hüllermeier, E.; et al. ChatGPT for good? On opportunities and challenges of large language models for education. Learn. Individ. Differ. 2023, 103, 102274. [CrossRef] AI 2023, 4 383 26. Schramowski, P.; Turan, C.; Andersen, N.; Rothkopf, C.A.; Kersting, K. Large pre-trained language models contain human-like biases of what is right and wrong to do. Nat. Mach. Intell. 2022, 4, 258–268. [CrossRef] 27. Zhou, J.; Chen, F.; Berry, A.; Reed, M.; Zhang, S.; Savage, S. A survey on ethical principles of AI and implementations. In Proceedings of the 2020 IEEE Symposium Series on Computational Intelligence (SSCI), Canberra, Australia, 1–4 December 2020; pp. 3010–3017. 28. Häußermann, J.J.; Lütge, C. Community-in-the-loop: Towards pluralistic value creation in AI, or—why AI needs business ethics. Ethics 2022, 2022, 1–22. [CrossRef] 29. Hosseini, M.; Horbach, S.P. Fighting reviewer fatigue or amplifying bias? Considerations and recommendations for use of ChatGPT and other Large Language Models in scholarly peer review. Res. Sq. 2023. Preprint. 30. Azaria, A. ChatGPT Usage and Limitations 2022. Preprint 2022. Available online: https://hal.science/hal-03913837/ (accessed on 2 February 2023). 31. Association, A.M. Definition of Marketing. Available online: https://www.ama.org/the-definition-of-marketing-what-is- marketing/ (accessed on 20 February 2023). 32. Ferrell, O.C.; Hartline, M.; Hochstein, B.W. Marketing Strategy, 8th ed.; Cengage Learning: Boston, MS, USA, 2021. 33. Verma, S.; Sharma, R.; Deb, S.; Maitra, D. Artificial intelligence in marketing: Systematic review and future research direction. Int. J. Inf. Manag. Data Insights 2021, 1, 100002. [CrossRef] 34. Oberoi, P. How Artificial Intelligence Is Impacting Marketing? In Encyclopedia of Data Science and Machine Learning; Wang, J., ed.; IGI Global: Hershey, PA, USA, 2023; pp. 606–618. [CrossRef] 35. Pitt, C.; Paschen, J.; Kietzmann, J.; Pitt, L.F.; Pala, E. Artificial intelligence, marketing, and the history of technology: Kranzberg’s laws as a conceptual lens. Australas. Mark. J. 2023, 31, 81–89. [CrossRef] 36. Zaman, K. Transformation of Marketing Decisions through Artificial Intelligence and Digital Marketing. J. Mark. Strateg. 2022, 4, 353–364. [CrossRef] 37. Lahoti, P.; Gummadi, K.P.; Weikum, G. Detecting and mitigating test-time failure risks via model-agnostic uncertainty learning. In Proceedings of the 2021 IEEE International Conference on Data Mining (ICDM), Auckland, New Zealand, 7–10 December 2021; pp. 1174–1179. 38. Yara, O.; Brazheyev, A.; Golovko, L.; Bashkatova, V. Legal Regulation of the Use of Artificial Intelligence: Problems and Development Prospects. Eur. J. Sustain. Dev. 2023, 10, 1. [CrossRef] 39. Sun, G.H.; Hoelscher, S.H. The ChatGPT Storm and What Faculty Can Do. Nurse Educ. 2023. Preprint. [CrossRef] 40. Stone, M.; Aravopoulou, E.; Ekinci, Y.; Evans, G.; Hobbs, M.; Labib, A.; Laughlin, P.; Machtynger, J.; Machtynger, L. Artificial Intelligence (AI) in Strategic Marketing Decision-Making: A research agenda. Bottom Line 2020, 33, 183–200. [CrossRef] 41. Krafft, M.; Sajtos, L.; Haenlein, M. Challenges and opportunities for marketing scholars in times of the fourth industrial revolution. J. Interact. Mark. 2020, 51, 1–8. [CrossRef] 42. Vlaci ˇ c, ´ B.; Corbo, L.; e Silva, S.C.; Dabic, ´ M. The evolving role of artificial intelligence in marketing: A review and research agenda. J. Bus. Res. 2021, 128, 187–203. [CrossRef] 43. Hermann, E. Leveraging artificial intelligence in marketing for social good—An ethical perspective. J. Bus. Ethics 2022, 179, 43–61. [CrossRef] 44. Grewal, D.; Hulland, J.; Kopalle, P.K.; Karahanna, E. The future of technology and marketing: A multidisciplinary perspective. J. Acad. Mark. Sci. 2020, 48, 1–8. [CrossRef] 45. Mogaji, E.; Soetan, T.O.; Kieu, T.A. The implications of artificial intelligence on the digital marketing of financial services to vulnerable customers. Australas. Mark. J. 2020, 29, 235–242. [CrossRef] 46. Paluch, S.; Wirtz, J. Editorial: Artificial Intelligence and Robots in the Service Encounter. J. Serv. Manag. Res. 2020, 4, 3–8. 47. Chowdhury, R.M.; Fernando, M. The Relationships of Empathy, Moral Identity and Cynicism with Consumers’ Ethical Beliefs: The Mediating Role of Moral Disengagement. J. Bus. Ethics 2014, 124, 677–694. [CrossRef] 48. Brand, J.; Israeli, A.; Ngwe, D. Using GPT for Market Research. SSRN. Preprint. 2023. [CrossRef] 49. Ma, L.; Sun, B. Machine learning and AI in marketing–Connecting computing power to human insights. Int. J. Res. Mark. 2020, 37, 481–504. [CrossRef] 50. Rivas, P.; Chelsi, C.; Nishit, N.; Ravula, L. Application-agnostic chatbot deployment considerations: A case study. In Proceedings of the 2019 International Conference on Computational Science and Computational Intelligence (CSCI), Las Vegas, NV, USA, 5–7 December 2019; pp. 361–365. 51. Akter, S.; McCarthy, G.; Sajib, S.; Michael, K.; Dwivedi, Y.K.; D’Ambra, J.; Shen, K.N. Algorithmic bias in data-driven innovation in the age of AI. Int. J. Inf. Manag. 2021, 60, 102387. [CrossRef] 52. De Bruyn, A.; Viswanathan, V.; Beh, Y.S.; Brock, J.K.U.; Von Wangenheim, F. Artificial intelligence and marketing: Pitfalls and opportunities. J. Interact. Mark. 2020, 51, 91–105. [CrossRef] 53. Wirtz, B.W.; Weyerer, J.C.; Sturm, B.J. The dark sides of artificial intelligence: An integrated AI governance framework for public administration. Int. J. Public Adm. 2020, 43, 818–829. [CrossRef] 54. Bogina, V.; Hartman, A.; Kuflik, T.; Shulner-Tal, A. Educating software and AI stakeholders about algorithmic fairness, accountability, transparency and ethics. Int. J. Artif. Intell. Educ. 2021, 32, 808–833 [CrossRef] 55. Kibbey, T. Transcriptivism: An ethical framework for modern linguistics. Proc. Linguist. Soc. Am. 2019, 4, 45. [CrossRef] 56. Etzioni, A.; Etzioni, O. Incorporating ethics into artificial intelligence. J. Ethics 2017, 21, 403–418. [CrossRef] AI 2023, 4 384 57. Agarwal, N. Artificial Intelligence and Marketing. Int. J. Soc. Sci. Econ. Res. 2022, 7. 58. Davenport, T.; Guha, A.; Grewal, D.; Bressgott, T. How artificial intelligence will change the future of marketing. J. Acad. Mark. Sci. 2020, 48, 24–42. [CrossRef] 59. Rivas, P. AI Orthopraxy: Towards a Framework for AI That Promotes Fairness. In Proceedings of the 2020 IEEE International Symposium on Technology and Society (ISTAS), Virtual, 12–15 November 2020; pp. 80–84. 60. Quinn, S.; Souppaya, M.; Cook, M.; Scarfone, K. National Checklist Program for IT Products: Guidelines for Checklist Users and Developers; Technical Report; National Institute of Standards and Technology: Gaithersburg, MD, USA, 2017. [CrossRef] 61. Sallam, M. The Utility of ChatGPT as an Example of Large Language Models in Healthcare Education, Research and Practice: Systematic Review on the Future Perspectives and Potential Limitations. medRxiv 2023. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Journal

AI – Multidisciplinary Digital Publishing Institute

Published: Apr 10, 2023

Keywords: marketing; ChatGPT; GPT technology; Chatbot; ethics; societal implications

There are no references for this article.