Avoid common mistakes on your manuscript.
1 Introduction
Tom Freston is credited with saying “Innovation is taking two things that exist and putting them together in a new way”. For a long time in history, it has been the prevailing assumption that artistic, creative tasks such as writing poems, creating software, designing fashion, and composing songs could only be performed by humans. This assumption has changed drastically with recent advances in artificial intelligence (AI) that can generate new content in ways that cannot be distinguished anymore from human craftsmanship.
The term generative AI refers to computational techniques that are capable of generating seemingly new, meaningful content such as text, images, or audio from training data. The widespread diffusion of this technology with examples such as Dall-E 2, GPT-4, and Copilot is currently revolutionizing the way we work and communicate with each other. Generative AI systems can not only be used for artistic purposes to create new text mimicking writers or new images mimicking illustrators, but they can and will assist humans as intelligent question-answering systems. Here, applications include information technology (IT) help desks where generative AI supports transitional knowledge work tasks and mundane needs such as cooking recipes and medical advice. Industry reports suggest that generative AI could raise global gross domestic product (GDP) by 7% and replace 300 million jobs of knowledge workers (Goldman Sachs 2023). Undoubtedly, this has drastic implications not only for the Business & Information Systems Engineering (BISE) community, where we will face revolutionary opportunities, but also challenges and risks that we need to tackle and manage to steer the technology and its use in a responsible and sustainable direction.
In this Catchword article, we provide a conceptualization of generative AI as an entity in socio-technical systems and provide examples of models, systems, and applications. Based on that, we introduce limitations of current generative AI and provide an agenda for BISE research. Previous papers discuss generative AI around specific methods such as language models (e.g., Teubner et al. 2023; Dwivedi et al. 2023; Schöbel et al. 2023) or specific applications such as marketing (e.g., Peres et al. 2023), innovation management (Burger et al. 2023), scholarly research (e.g., Susarla et al. 2023; Davison et al. 2023), and education (e.g., Kasneci et al. 2023; Gimpel et al. 2023). Different from these works, we focus on generative AI in the context of information systems, and, to this end, we discuss several opportunities and challenges that are unique to the BISE community and make suggestions for impactful directions for BISE research.
2 Conceptualization
2.1 Mathematical Principles of Generative AI
Generative AI is primarily based on generative modeling, which has distinctive mathematical differences from discriminative modeling (Ng and Jordan 2001) often used in data-driven decision support. In general, discriminative modeling tries to separate data points X into different classes Y by learning decision boundaries between them (e.g., in classification tasks with \(Y \in \{ 0, 1 \}\)). In contrast to that, generative modeling aims to infer some actual data distribution. Examples can be the joint probability distribution P(X, Y) of both the inputs and the outputs or P(Y), but where Y is typically from some high-dimensional space. By doing so, a generative model offers the ability to produce new synthetic samples (e.g., generate new observation-target-pairs (X, Y) or new observations X given a target value Y) (Bishop 2006).
Building upon the above, a generative AI model refers to generative modeling that is instantiated with a machine learning architecture (e.g., a deep neural network) and, therefore, can create new data samples based on learned patterns.Footnote 1 Further, a generative AI system encompasses the entire infrastructure, including the model, data processing, and user interface components. The model serves as the core component of the system, which facilitates interaction and application within a broader context. Lastly, generative AI applications refer to the practical use cases and implementations of these systems, such as search engine optimization (SEO) content generation or code generation that solve real-world problems and drive innovation across various domains. Figure 1 shows a systematization of generative AI across selected data modalities (e.g., text, image, and audio) and the model-, system-, and application-level perspectives, which we detail in the following section.
A model-, system-, and application-level view on generative AI
Note that the modalities in Fig. 1 are neither complete nor entirely distinctive and can be detailed further. In addition, many unique use cases such as, for example, modeling functional properties of proteins (Unsal et al. 2022) can be represented in another modality such as text.
2.2 A Model-, System-, and Application-Level View of Generative AI
2.2.1 Model-Level View
A generative AI model is a type of machine learning architecture that uses AI algorithms to create novel data instances, drawing upon the patterns and relationships observed in the training data. A generative AI model is of critically central yet incomplete nature, as it requires further fine-tuning to specific tasks through systems and applications.
Deep neural networks are particularly well suited for the purpose of data generation, especially as deep neural networks can be designed using different architectures to model different data types (Janiesch et al. 2021; Kraus et al. 2020), for example, sequential data such as human language or spatial data such as images. Table 1 presents an overview of the underlying concepts and model architectures that are common in the context of generative AI, such as diffusion probabilistic models for text-to-image generation or the transformer architecture and (large) language models (LLMs) for text generation. GPT (short for generative pre-trained transformer), for example, represents a popular family of LLMs, used for text generation, for instance, in the conversational agent ChatGPT.
Large generative AI models that can model output in and across specific domains or specific data types in a comprehensive and versatile manner are oftentimes also called foundation models (Bommasani et al. 2021) powered by LLMs (Park et al. 2021). With generative AI, a human uses prompts to engage with an AI system to create content, and the AI then interprets the human’s intentions and provides feedback to presuppose further prompts. At first glance, this seems to follow a delegation pattern as well. Yet, the subsequent process does not, as the output of the AI can be suggestive to the other and will inform their further involvement directly or subconsciously. Thus, the process of creation rather follows a co-creation pattern, that is, the practice of collaborating in different roles to align and offer diverse insights to guide a design process (Ramaswamy and Ozcan 2018). Using the lens of agentic AI artifacts, initiation is not limited to humans.
The abovementioned interactions also impact our current understanding of hybrid intelligence as the integration of humans and AI, leveraging the unique strengths of both. Hybrid intelligence argues to address the limitations of each intelligence type by combining human intuition, creativity, and empathy with the computational power, accuracy, and scalability of AI systems to achieve enhanced decision-making and problem-solving capabilities (Dellermann et al. 2019). With generative AI and the AI’s capability to co-create, the understanding of what constitutes this collective intelligence begins to shift. Hence, novel human-AI interaction models and patterns may become necessary to explain and guide the behavior of humans and AI systems to enable effective and efficient use in AI applications on the one hand and, on the other hand, to ensure envelopment of AI agency and reach (Asatiani et al. 2021).
On a theoretical level, this shift in human-computer or rather human-AI interaction fuels another important observation: The theory of mind is an established theoretical lens in psychology to describe the cognitive ability of individuals to understand and predict the mental states, emotions, and intentions of others (Carlson et al. 2013; Baron-Cohen 1997; Gray et al. 2007). This skill is crucial for social interactions, as it facilitates empathy and allows for effective communication. Moreover, conferring a mind to an AI system can substantially drive usage intensity (Hartmann et al. 2023a). The development of a theory of mind in humans is unconscious and evolves throughout an individual’s life. The more natural AI systems become in terms of their interface and output, the more a theory of mind for human-computer interactions becomes necessary. Research is already investigating how AI systems can become theory-of-mind-aware to better understand their human counterpart (Rabinowitz et al. 2018; Çelikok et al. 2019). However, current AI systems hardly offer any cues for interactions. Thus, humans are rather void of a theory to explain their understanding of intelligent behavior by AI systems, which becomes even more important in a co-creation environment that does not follow a task delegation pattern. A theory of the artificial mind that explains how individuals perceive and assume the states and rationale of AI systems to better collaborate with them may alleviate some of these concerns.
3 Limitations of Current Generative AI
In the following, we discuss four salient boundaries of generative AI that, we argue, are important limitations in real-world applications. The following limitations are of technical nature in that they refer to how current generative AI models make inferences, and, hence, the limitations arise at the model level. Because of this, it is likely that limitations will persist in the long run, with system- and application-level implications.
Incorrect outputs. Generative AI models may produce output with errors. This is owed to the underlying nature of machine learning models relying on probabilistic algorithms for making inferences. For example, generative AI models generate the most probable response to a prompt, not necessarily the correct response. As such, challenges arise as, by now, outputs are indistinguishable from authentic content and may present misinformation or deceive users (Spitale et al. 2023). In LLMs, this problem in emergent behavior is called hallucination (Ji et al. 2023), which refers to mistakes in the generated text that are semantically or syntactically plausible but are actually nonsensical or incorrect. In other words, the generative AI model produces content that is not based on any facts or evidence, but rather on its own assumptions or biases. Moreover, the output of generative AI, especially that of LLMs, is typically not easily verifiable.
The correctness of generative AI models is highly dependent on the quality of training data and the according learning process. Generative AI systems and applications can implement correctness checks to inhibit certain outputs. Yet, due to the black-box nature of state-of-the-art AI models (Rai 2020), the usage of such systems critically hinges on users’ trust in reliable outputs. The closed source of commercial off-the-shelf generative AI systems aggravates this fact and prohibits further tuning and re-training of the models. One solution for addressing the downstream implications of incorrect outputs is to use generative AI to produce explanations or references, which can then be verified by users. However, such explanations are again probabilistic and thus subject to errors; nevertheless, they may help users in their judgment and decision-making when to accept outputs of generative AI and when not.
Bias and fairness. Societal biases permeate everyday human-generated content (Eskreis-Winkler and Fishbach 2022). The unbiasedness of vanilla generative AI is very much dependent on the quality of training data and the alignment process. Training deep learning models on biased data can amplify human biases, replicate toxic language, or perpetuate stereotypes of gender, sexual orientation, political leaning, or religion (e.g., Caliskan et al. 2017; Hartmann et al. 2020). For example, the carbon emission for training a generative AI model such as GPT-3 was estimated to have produced the equivalent of 552 t \(\hbox {CO}_2\) and thus amounts to the annual \(\hbox {CO}_2\) emissions of several dozens of households (Khan 2021). Owing to this, there are ongoing efforts in AI research to make the development and deployment of AI algorithms more carbon-friendly, through more efficient training algorithms, through compressing the size of neural network architectures, and through optimized hardware (Schwartz et al. 2020).
4 Implications and Future Directions for the BISE Community
In this section, we draw a number of implications and future research directions which, on the one hand, are of direct relevance to the BISE community as an application-oriented, socio-technical research discipline and, on the other hand, offer numerous research opportunities, especially for BISE researchers due to their interdisciplinary background. We organize our considerations according to the individual departments of the BISE journal (see Table 2 for an overview of exemplary research questions).
4.1 Business Process Management
Generative AI will have a strong impact on the field of Business Process Management (BPM) as it can assist in automating routine tasks, improving customer and employee satisfaction, and revealing process innovation opportunities (Beverungen et al. 2021), especially in creative processes (Haase and Hanel 2021) will benefit as formerly handcrafted processing rules can not only be replaced, but entirely new types of automation can be enabled by retrofitting and thus intelligentizing legacy software. In the long run, we also see a large potential to support the phase of business process execution in traditional BPM. Specifically, we anticipate the development of a new generation of process guidance systems. While traditional system designs are based on static and manually-crafted knowledge bases (Morana et al. 2019), more dynamic and adaptive systems are feasible on the basis of large enterprise-wide trained language models. Such systems could improve knowledge retrieval tasks from a wide variety of heterogeneous sources, including manuals, handbooks, e-mails, wikis, job descriptions, etc. This opens up new avenues of research into how unstructured and distributed organizational knowledge can be incorporated into intelligent process guidance systems.
4.2 Decision Analytics and Data Science
Despite the huge progress in recent years, several analytical and technical questions around the development of generative AI have yet to be solved. One open question relates to how generative AI can be effectively customized for domain-specific applications and thus improve performance through higher degrees of contextualization. For example, novel and scalable techniques are needed to customize conversational agents based on generative AI for applications in medicine or finance. This will be crucial in practice to solve specific BISE-related tasks where customization may bring additional performance gains. Novel techniques for customization must be designed in a way that ensures the safety of proprietary data and prevents the data from being disclosed. Moreover, new frameworks are needed for prompt engineering that are designed from a user-centered lens and thus promote interpretability and usability.
Another important research direction is to improve the reliability of generative AI systems. For example, algorithmic solutions are needed on how generative AI can detect and mitigate hallucination. In addition to algorithmic solutions, more effort is also needed to develop user-centered solutions, that is, how users can reduce the risk of falling for incorrect outcomes, for example, by develo** better ways how outputs can be verified (e.g., by offering additional explanations or references).
Finally, questions arise about how generative AI can natively support decision analytics and data science projects by closing the gap between modeling experts and domain users (Zschech et al. 2020). For instance, it is commonly known that many AI models used in business analytics are difficult to understand by non-experts (cf. Senoner et al. 2022). As a remedy, generative AI could be used to generate descriptions that explain the logic of business analytics models and thus make the decision logic more intelligible. One promising direction could be, for example, to use generative AI for translating post hoc explanations derived from approaches like SHAP or LIME into more intuitive textual descriptions or generate user-friendly descriptions of models that are intrinsically interpretable (Slack et al. 2023; Zilker et al. 2023).
4.3 Digital Business Management and Digital Leadership
Generative AI has great potential to contribute to different types of value creation mechanisms, including knowledge creation, task augmentation, and autonomous agency. However, this also requires the necessary organizational capabilities and conditions, where further research is needed to examine these ingredients more closely for the context of generative AI to steer the technological possibilities in a successful direction (Shollo et al. 2022).
That is, generative AI will lead to the development of new business ideas, unseen product and service innovations, and ultimately to the emergence of completely new business models. At the same time, it will also have a strong impact on intra-organizational aspects, such as work patterns, organizational structures, leadership models, and management practices. In this regard, we see that AI-based assistant systems previously centered around desktop automation taking over more and more routine tasks such as event management, resource allocation, and social media account management to free up even more human capacity (Maedche et al. 2019). Further, in algorithmic management (Benlian et al. 2022; Cameron et al. 2023), it should be examined how existing theories and frameworks need to be contextualized or fundamentally extended in light of the increasingly powerful capabilities of generative AI.
However, there are not only implications at the management level. The future of work is very likely to change at all levels of an organization (Feuerriegel et al. 2022). Due to the multi-modality of generative AI models, it is conceivable that employees will work increasingly via smart, speech-based interfaces, whereby the formulation of prompts and the evaluation of their results could become a key activity. Against this background, it is worth investigating which new competencies are required to handle this emerging technology (cf. Debortoli et al. 2014) and which entirely new job profiles, such as prompt engineers, may evolve in the near future (Strobelt et al. 2023).
Generative AI is also expected to fundamentally reform the way organizations manage, maintain, and share knowledge. Referring to the sketched vision of a new process guidance system in Sect. 4.1, we anticipate a number of new opportunities for digital knowledge management, among others automated knowledge discovery based on large amounts of unstructured distributed data (e.g., identification of new product combinations), improved knowledge sharing by automating the process of creating, summarizing, and disseminating content (e.g., automated creation of wikis and FAQs in different languages), and personalized knowledge delivery to individual employees based on their specific needs and preferences (e.g., recommendations for specific training material).
4.4 Economics of Information Systems
Generative AI will have significant economic implications across various industries and markets. Generative AI can increase efficiency and productivity by automating many tasks that were previously performed by humans, such as content creation, customer service, code generation, etc. This can reduce costs and open up new opportunities for growth and innovation (Eloundou et al. 2023). For example, AI-based translation between different languages is responsible for significant economic gains (Brynjolfsson et al. 2019). The BISE community can contribute by providing quantification through rigorous causal evidence. Given the velocity of AI research, it may be necessary to take a more abstract problem view instead of a concrete tool view. For example, BISE research could run field experiments to compare programmers with and without AI support and thereby assess whether generative AI systems for coding can improve the speed and quality of code development. Similarly, researchers could test whether generative AI will make artists more creative as they can more easily create new content. A similar pattern was previously observed for AlphaGo, which has led humans to become better players in the board game Go (Shin et al. 2023).
Generative AI is likely to transform the industry as a whole. This may hold true in the case of platforms that make user-generated content available (e.g., shutterstock.com, pixabay.com, stackoverflow.com), which may be replaced by generative AI systems. Here, further research questions arise as to whether the use of generative AI can lead to a competitive advantage and how generative AI changes competition. For example, what are the economic implications if generative AI is developed as open-source vs. closed-source systems? In this regard, a salient success factor for the development of conversational agents based on generative AI (e.g., ChatGPT) are data from user interactions through dialogues and feedback on whether the dialog was helpful. Hence, the value of such interaction data is poorly understood and what it means if such data are only available to a few Big Tech companies.
The digital transformation from generative AI also poses challenges and opportunities for economic policy. It may affect future work patterns and, indirectly, worker capability via restructured learning mechanisms. It may also affect content sharing and distribution and, hence, have non-trivial implications on the exploitation and protection of intellectual properties. On top of that, a growing concentration of power over AI innovation in the hands of a few companies may result in a monopoly of AI capabilities and hamper future innovation, fair competition, scientific progress, and thus welfare and human development at large. All of these future impacts are important to understand and provide meaningful directions for sha** economic policy.
4.5 Enterprise Modeling and Enterprise Engineering
Enterprise models are important artifacts for capturing insights into the core components and structures of an organization, including business processes, resources, information flows, and IT systems (Vernadat 2020). A major drawback of traditional enterprise models is that they are static and may not provide the level of abstraction that is required by the end user. Likewise, their construction and maintenance are time-consuming and expensive and require manual effort and human expertise (Silva et al. 2021). With generative AI, we see a large potential that many of these limitations can be addressed by generative AI as assistive technology (Sandkuhl et al. 2018), for example by automatically creating and updating enterprise models at different levels of abstraction or generating multi-modal representations.
First empirical results suggest that generative AI is able to generate useful conceptual models based on textual problem descriptions. Fill et al. (2023) show that ER, BPMN, UML, and Heraklit models can not only be generated with very high to perfect accuracy from textual descriptions, but they also explored the interpretation of existing models and received good results. In the near future, we expect more research that deals with the development, evaluation, and application of more advanced approaches. Specifically, we expect that learned representations of enterprise models can be transformed into more application-specific formats and can either be enriched with further details or reduced to the essential content.
Against this background, the concept of “digital twins”, virtual representations of enterprise assets, may experience new accentuation and extensions (Dietz and Pernul 2020). Especially, in the public sector, where most organizational assets are non-tangible in the form of defined services, specified procedures, legal texts, manuals, and organizational charts, generative AI can play a crucial role in digitally mirroring and managing such assets along their lifecycles. Similar benefits could be explored with physical assets in Industry 4.0 environments (Lasi et al. 2014).
In enterprise engineering, the role of generative AI systems in existing as well as newly emerging IT landscapes to support the business goals and strategies of an organization gives rise to numerous opportunities (e.g., in office solutions, customer relationship management and business analytics applications, knowledge management systems, etc.). Generative AI systems have the potential to evolve into core enterprise applications that can either be hosted on-premise or rented in the cloud. Unsanctioned use bears the risk that third-party applications will be used for job-related tasks without explicit approval or even knowledge of the organization. This phenomenon is commonly known as shadow IT and theories and frameworks have been proposed to explain this phenomenon, as well as recommending actions and policies to mitigate associated risks (cf. Haag and Eckhardt 2017; Klotz et al. 2022). In the light of generative AI, however, such approaches have to be revisited for their applicability and effectiveness and, if necessary, need to be extended. Nevertheless, this situation also offers the potential to explore and design new approaches for more effective API management (e.g., including novel app store solutions, privacy and security mechanisms, service level definitions, pricing, and licensing models) so that generative AI solutions can be smoothly integrated into existing enterprise IT infrastructures without risking any unauthorized use and confidentiality breaches.
4.6 Human Computer Interaction and Social Computing
Salient behavioral questions related to the interactions between humans and generative AI systems are still unanswered. Examples are related to the perception, acceptance, adoption, and trust of systems using generative AI. A study found that news was believed less if generated by generative AI instead of humans (Longoni et al. 2022) and another found that there is a replicant effect (Jakesch et al. 2019). Such behavior is likely to be context-specific and will vary by other antecedents highlighting the need for a principled theoretical foundation to build successful generative AI systems. The BISE community is well positioned to develop rigorous design recommendations.
Further, generative AI is a key enabler for develo** high-quality interfaces for information systems based on natural language that promote usability and accessibility. For example, such interfaces will not only make interactions more intuitive but will also facilitate people with disabilities. Generative AI is likely to increase the “degree of intelligence” of user assistance systems. However, the design of effective interactions must also be considered when increasing the degree of intelligence (Maedche et al. 2016). Similarly, generative AI will undoubtedly have an impact on (computer-mediated) communication and collaboration, such as within companies. For example, generative AI can create optimized content for social media, emails, and reports. It can also help to improve the onboarding of new employees by creating personalized and interactive training materials. It can also enhance collaboration within teams by providing creative and intelligence conservation agents that suggest, summarize, and synthesize information based on the context of the team (e.g., automated meeting notes).
Several applications and research opportunities are related to the use of generative AI in marketing and, especially, e-commerce. It is expected that generative AI can automate the creation of personalized marketing content, for instance, different sales slogans for introverts vs. extroverts (Matz et al. 2017) or other personality traits as personalized marketing content is more effective than a one-content-fits-all approach (Matz et al. 3). As such, design principles can focus on how generative AI systems can be made explainable to enable interpretability, understanding, and trust; how they can be designed reliable to avoid discrimination effects or privacy issues; and how they can be built more energy efficient to promote environmental sustainability (cf. Schoormann et al. 2023b). While a lot of research is already being conducted in technology-oriented disciplines such as computer science, the BISE community can add its strength by looking at design aspects through a socio-technical lens, involving individuals, teams, organizations, and societal groups in design activities, and thereby driving the field forward with new insights from a human–machine perspective (Maedche et al. 2019).
Further, we see great potential that generative AI can be leveraged to improve current practices in design science research projects when constructing novel IT artifacts (see Hevner et al. 2019). Here, one of the biggest potentials could lie in the support of knowledge retrieval tasks. Currently, design knowledge in the form of design requirements, design principles, and design features is often only available in encapsulated written papers or implicitly embedded in instantiated artifacts. Generative AI has the potential to extract such design knowledge that is spread over a broad body of interdisciplinary research and make it available in a collective form for scholars and practitioners. This could also overcome the limitation that design knowledge is currently rarely reused, which hampers the fundamental idea of knowledge accumulation in design science research (Schoormann et al. 2021).
Besides engineering actual systems and applications, the BISE community should also investigate how generative AI can be used to support creativity-based tasks when initiating new design projects. In this regard, a promising direction could be to incorporate generative AI in design thinking and similar methodologies to combine human creativity with computational creativity (Hawlitschek 2023). This may support different phases and steps of innovation projects, such as idea generation, user needs elicitation, prototy**, design evaluation, and design automation, in which different types of generative AI models and systems could be used and combined with each other to form applications for creative industries (e.g., generated user stories with textual descriptions, visual mock-ups for user interfaces, and quick software prototypes for proofs-of-concept). If generative AI is used to co-create innovative outcomes, it may also enable better reflection of the different design activities to ensure the necessary learning (Schoormann et al. 2023a).
5 Conclusion
Generative AI is a branch of AI that can create new content such as texts, images, or audio that increasingly often cannot be distinguished anymore from human craftsmanship. For this reason, generative AI has the potential to transform domains and industries that rely on creativity, innovation, and knowledge processing. In particular, it enables new applications that were previously impossible or impractical for automation, such as realistic virtual assistants, personalized education and service, and digital art. As such, generative AI has substantial implications for BISE practitioners and scholars as an interdisciplinary research community. In our Catchword article, we offered a conceptualization of the principles of generative AI along a model-, system-, and application-level view as well as a social-technical view and described limitations of current generative AI. Ultimately, we provided an impactful research agenda for the BISE community and thereby highlight the manifold affordances that generative AI offers through the lens of the BISE discipline.
Notes
It should be noted, however, that advanced generative AI models are often not based on a single modeling principle or learning mechanism, but combine different approaches. For example, language models from the GPT family first apply a generative pre-training stage to capture the distribution of language data using a language modeling objective, while downstream systems typically then apply a discriminative fine-tuning stage to adapt the model parameters to specific tasks (e.g., document classification, question answering). Similarly, ChatGPT combines techniques from generative modeling together with discriminatory modeling and reinforcement learning (see Fig. 2).
See https://github.com/huggingface/olm-datasets (accessed 25 Aug 2023) for a script that enables users to pull up-to-date data from the web for training online language models, for instance, from Common Crawl and Wikipedia.
References
Agostinelli A, Denk TI, Borsos Z, Engel J, Verzetti M, Caillon A, Huang Q, Jansen A, Roberts A, Tagliasacchi M, et al (2023) MusicLM: generating music from text. ar**v:2301.11325
Asatiani A, Malo P, Nagbøl PR, Penttinen E, Rinta-Kahila T, Salovaara A (2021) Sociotechnical envelopment of artificial intelligence: an approach to organizational deployment of inscrutable artificial intelligence systems. J Assoc Inf Syst 22(2):8
Baird A, Maru** LM (2021) The next generation of research on IS use: a theoretical framework of delegation to and from agentic IS artifacts. MIS Q 45(1):315–341
Baron-Cohen S (1997) Mindblindness: an essay on autism and theory of mind. MIT Press, Cambridge
Benlian A, Wiener M, Cram WA, Krasnova H, Maedche A, Möhlmann M, Recker J, Remus U (2022) Algorithmic management. Bus Inf Syst Eng 64(6):825–839. https://doi.org/10.1007/s12599-022-00764-w
Berente N, Gu B, Recker J, Santhanam R (2021) Special issue editor’s comments: managing artificial intelligence. MIS Q 45(3):1433–1450
Beverungen D, Buijs JCAM, Becker J, Di Ciccio C, van der Aalst WMP, Bartelheimer C, vom Brocke J, Comuzzi M, Kraume K, Leopold H, Matzner M, Mendling J, Ogonek N, Post T, Resinas M, Revoredo K, del Río-Ortega A, La Rosa M, Santoro FM, Solti A, Song M, Stein A, Stierle M, Wolf V (2021) Seven paradoxes of business process management in a hyper-connected world. Bus Inf Syst Eng 63(2):145–156. https://doi.org/10.1007/s12599-020-00646-z
Birhane A, Prabhu VU, Kahembwe E (2021) Multimodal datasets: misogyny, pornography, and malignant stereotypes. ar**v:2110.01963
Bishop C (2006) Pattern recognition and machine learning. Springer, New York
Bommasani R, Hudson DA, Adeli E, Altman R, Arora S, von Arx S, Bernstein MS, Bohg J, Bosselut A, Brunskill E, Brynjolfsson E, Buch S, Card D, Castellon R, Chatterji NS, Chen AS, Creel KA, Davis J, Demszky D, Donahue C, Doumbouya M, Durmus E, Ermon S, Etchemendy J, Ethayarajh K, Fei-Fei L, Finn C, Gale T, Gillespie LE, Goel K, Goodman ND, Grossman S, Guha N, Hashimoto T, Henderson P, Hewitt J, Ho DE, Hong J, Hsu K, Huang J, Icard TF, Jain S, Jurafsky D, Kalluri P, Karamcheti S, Keeling G, Khani F, Khattab O, Koh PW, Krass MS, Krishna R, Kuditipudi R, Kumar A, Ladhak F, Lee M, Lee T, Leskovec J, Levent I, Li XL, Li X, Ma T, Malik A, Manning CD, Mirchandani SP, Mitchell E, Munyikwa Z, Nair S, Narayan A, Narayanan D, Newman B, Nie A, Niebles JC, Nilforoshan H, Nyarko JF, Ogut G, Orr L, Papadimitriou I, Park JS, Piech C, Portelance E, Potts C, Raghunathan A, Reich R, Ren H, Rong F, Roohani YH, Ruiz C, Ryan J, R’e C, Sadigh D, Sagawa S, Santhanam K, Shih A, Srinivasan KP, Tamkin A, Taori R, Thomas AW, Tramèr F, Wang RE, Wang W, Wu B, Wu J, Wu Y, **e SM, Yasunaga M, You J, Zaharia MA, Zhang M, Zhang T, Zhang X, Zhang Y, Zheng L, Zhou K, Liang P (2021) On the opportunities and risks of foundation models. ar**v:2108.07258https://doi.org/10.48550/ar**v.2108.07258
Brand J, Israeli A, Ngwe D (2023) Using GPT for market research. SSRN 4395751
Brown T, Mann B, Ryder N, Subbiah M, Kaplan JD, Dhariwal P, Neelakantan A, Shyam P, Sastry G, Askell A et al (2020) Language models are few-shot learners. Adv Neural Inf Process Syst 33:1877–1901
Brynjolfsson E, Hui X, Liu M (2019) Does machine translation affect international trade? Evidence from a large digital platform. Manag Sci 65(12):5449–5460
Burger B, Kanbach DK, Kraus S, Breier M, Corvello V (2023) On the use of AI-based tools like ChatGPT to support management research. Europ J Innov Manag 26(7):233–241. https://doi.org/10.1108/EJIM-02-2023-0156
Busch K, Rochlitzer1 A, Sola D, Leopold H (2023) Just tell me: Prompt engineering in business process management. ar**v:2304.07183
Caliskan A, Bryson JJ, Narayanan A (2017) Semantics derived automatically from language corpora contain human-like biases. Sci 356(6334):183–186
Cameron L, Lamers L, Leicht-Deobald U, Lutz C, Meijerink J, Möhlmann M (2023) Algorithmic management: its implications for information systems research. Commun AIS 52(1):518–537. https://doi.org/10.17705/1CAIS.05221
Carlson SM, Koenig MA, Harms MB (2013) Theory of mind. WIREs Cogn Sci 4:391–402
Çelikok MM, Peltola T, Daee P, Kaski S (2019) Interactive AI with a theory of mind. In: ACM CHI 2019 workshop: computational modeling in human-computer interaction, vol 80, pp 4215–4224
Chen L, Zaharia M, Zou J (2023) How is chatgpt’s behavior changing over time? ar**v:2307.09009
Chen M, Tworek J, Jun H, Yuan Q, Pinto HPdO, Kaplan J, Edwards H, Burda Y, Joseph N, Brockman G, et al (2021) Evaluating large language models trained on code. ar**v:2107.03374
Chiang T (2023) ChatGPT is a blurry JPEG of the web. https://www.newyorker.com/tech/annals-of-technology/chatgpt-is-a-blurry-jpeg-of-the-web, accessed 25 Aug 2023
Davison RM, Laumer S, Tarafdar M, Wong LHM (2023) ISJ editorial: pickled eggs: generative AI as research assistant or co-author? Inf Syst J Early View. https://doi.org/10.1111/isj.12455
De-Arteaga M, Feuerriegel S, Saar-Tsechansky M (2022) Algorithmic fairness in business analytics: directions for research and practice. Prod Oper Manag 31(10):3749–3770
Debortoli S, Müller O, vom Brocke J (2014) Comparing business intelligence and big data skills. Bus Inf Syst Eng 6(5):289–300. https://doi.org/10.1007/s12599-014-0344-2
Dellermann D, Ebel P, Söllner M, Leimeister JM (2019) Hybrid intelligence. Bus Inf Syst Eng 61(5):637–643. https://doi.org/10.1007/s12599-019-00595-2
Devlin J, Chang MW, Lee K, Toutanova K (2018) BERT: Pre-training of deep bidirectional transformers for language understanding. ar**v:1810.04805
Diederich S, Brendel AB, Kolbe LM (2020) Designing anthropomorphic enterprise conversational agents. Bus Inf Syst Eng 62(3):193–209
Dietz M, Pernul G (2020) Digital twin: empowering enterprises towards a system-of-systems approach. Bus Inf Syst Eng 62(2):179–184. https://doi.org/10.1007/s12599-019-00624-0
Dolata M, Feuerriegel S, Schwabe G (2022) A sociotechnical view of algorithmic fairness. Inf Syst J 32(4):754–818
van Dun C, Moder L, Kratsch W, Röglinger M (2023) ProcessGAN: supporting the creation of business process improvement ideas through generative machine learning. Decis Support Syst 165(113):880. https://doi.org/10.1016/j.dss.2022.113880
Dwivedi YK, Kshetri N, Hughes L, Slade EL, Jeyaraj A, Kar AK, Baabdullah AM, Koohang A, Raghavan V, Ahuja M et al (2023) “So what if ChatGPT wrote it?’’ Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policy. Int J Inf Manag 71(102):642
Eloundou T, Manning S, Mishkin P, Rock D (2023) GPTs are GPTs: an early look at the labor market impact potential of large language models. arxiv:2303.10130, accessed 03 April 2023
Eskreis-Winkler L, Fishbach A (2022) Surprised elaboration: when white men get longer sentences. J Personal Soc Psychol 123:941–956
Ferrara E (2023) Should ChatGPT be biased? Challenges and risks of bias in large language models. ar**v:2304.03738
Feuerriegel S, Dolata M, Schwabe G (2020) Fair AI: challenges and opportunities. Bus Inf Syst Eng 62:379–384
Feuerriegel S, Shrestha YR, von Krogh G, Zhang C (2022) Bringing artificial intelligence to business management. Nat Machine Intell 4(7):611–613
Fill HG, Fettke P, Köpke J (2023) Conceptual modeling and large language models: impressions from first experiments with ChatGPT. EMISAJ 18(3):1–15. https://doi.org/10.18417/emisa.18.3
Ganguli D, Askell A, Schiefer N, Liao T, Lukošiūtė K, Chen A, Goldie A, Mirhoseini A, Olsson C, Hernandez D, et al (2023) The capacity for moral self-correction in large language models. ar**v:2302.07459
Garcia T (2023) David Guetta replicated Eminem’s voice in a song using artificial intelligence. https://variety.com/2023/music/news/david-guetta-eminem-artificial-intelligence-1235516924/, accessed 25 Aug 2023
Gilardi F, Alizadeh M, Kubli M (2023) ChatGPT outperforms crowd-workers for text-annotation tasks. ar**v:2303.15056
Gimpel H, Hall K, Decker S, Eymann T, Lämmermann L, Mädche A, Röglinger M, Ruiner C, Schoch M, Schoop M, et al (2023) Unlocking the power of generative ai models and systems such as GPT-4 and ChatGPT for higher education. https://digital.uni-hohenheim.de/fileadmin/einrichtungen/digital/Generative_AI_and_ChatGPT_in_Higher_Education.pdf, accessed 25 Aug 2023
Goldman Sachs (2023) Generative AI could raise global GDP by 7%. https://www.goldmansachs.com/insights/pages/generative-ai-could-raise-global-gdp-by-7-percent.html
Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial nets. Adv Neural Inf Process Syst 27:2672–2680
Gray HM, Gray K, Wegner DM (2007) Dimensions of mind perception. Sci 315(5812):619–619
Grisold T, Groß S, Stelzl K, vom Brocke J, Mendling J, Röglinger M, Rosemann M (2022) The five diamond method for explorative business process management. Bus Inf Syst Eng 64(2):149–166. https://doi.org/10.1007/s12599-021-00703-1
Haag S, Eckhardt A (2017) Shadow IT. Bus Inf Syst Eng 59(6):469–473. https://doi.org/10.1007/s12599-017-0497-x
Haase J, Hanel PHP (2023) Artificial muses: generative artificial intelligence chatbots have risen to human-level creativity. ar**v:2303.12003
Hartmann J, Bergner A, Hildebrand C (2023a) MindMiner: uncovering linguistic markers of mind perception as a new lens to understand consumer-smart object relationships. J Consum Psychol. https://doi.org/10.1002/jcpy.1381
Hartmann J, Schwenzow J, Witte M (2023b) The political ideology of conversational AI: converging evidence on ChatGPT’s pro-environmental, left-libertarian orientation. ar**v:2301.01768
Hawlitschek F (2023) Interview with Samuel Tschepe on “Quo vadis design thinking?’’. Bus Inf Syst Eng 65(2):223–228. https://doi.org/10.1007/s12599-023-00792-0
Herm LV, Janiesch C, Reijers HA, Seubert F (2021) From symbolic RPA to intelligent RPA: challenges for develo** and operating intelligent software robots. In: International conference on business process management, pp 289–305
Hevner A, vom Brocke J, Maedche A (2019) Roles of digital innovation in design science research. Bus Inf Syst Eng 61(1):3–8. https://doi.org/10.1007/s12599-018-0571-z
Ho J, Jain A, Abbeel P (2020) Denoising diffusion probabilistic models. Adv Neural Inf Process Syst 33:6840–6851
Jakesch M, French M, Ma X, Hancock JT, Naaman M (2019) AI-mediated communication: how the perception that profile text was written by AI affects trustworthiness. In: Conference on human factors in computing systems (CHI)
Jakesch M, Hancock JT, Naaman M (2023) Human heuristics for AI-generated language are flawed. Proc Natl Acad Sci 120(11):e2208839
Janiesch C, Zschech P, Heinrich K (2021) Machine learning and deep learning. Electron Market 31(3):685–695. https://doi.org/10.1007/s12525-021-00475-2
Ji Z, Lee N, Frieske R, Yu T, Su D, Xu Y, Ishii E, Bang YJ, Madotto A, Fung P (2023) Survey of hallucination in natural language generation. ACM Comput Surv 55(12):1–38
Kasneci E, Seßler K, Küchemann S, Bannert M, Dementieva D, Fischer F, Gasser U, Groh G, Günnemann S, Hüllermeier E et al (2023) ChatGPT for good? On opportunities and challenges of large language models for education. Learn Individ Differ 103(102):274
Kecht C, Egger A, Kratsch W, Röglinger M (2023) Quantifying chatbots’ ability to learn business processes. Inf Syst 113(102):176. https://doi.org/10.1016/j.is.2023.102176
Khan J (2021) AI’s carbon footprint is big, but easy to reduce, Google researchers say. Fortune
Kingma DP, Welling M (2013) Auto-encoding variational Bayes. https://doi.org/10.48550/ar**v.1312.6114
Klotz S, Westner M, Strahringer S (2022) Critical success factors of business-managed IT: it takes two to tango. Inf Syst Manag 39(3):220–240
Kraus M, Feuerriegel S, Oztekin A (2020) Deep learning in business analytics and operations research: models, applications and managerial implications. Europ J Oper Res 281(3):628–641. https://doi.org/10.1016/j.ejor.2019.09.018
Kreps S, McCain RM, Brundage M (2022) All the news that’s fit to fabricate: AI-generated text as a tool of media misinformation. J Exp Polit Sci 9(1):104–117
Krügel S, Ostermaier A, Uhl M (2023) ChatGPT’s inconsistent moral advice influences users’ judgment. Sci Report 13(1):4569
Lasi H, Fettke P, Kemper HG, Feld T, Hoffmann M (2014) Industry 4.0. Bus Inf Syst Eng 6(4):239–242. https://doi.org/10.1007/s12599-014-0334-4
Li Y, Choi D, Chung J, Kushman N, Schrittwieser J, Leblond R, Eccles T, Keeling J, Gimeno F, Dal Lago A et al (2022) Competition-level code generation with alphacode. Science 378(6624):1092–1097
Liu P, Yuan W, Fu J, Jiang Z, Hayashi H, Neubig G (2023) Pre-train, prompt, and predict: a systematic survey of prompting methods in natural language processing. ACM Comput Surv 55(9):1–35
Longoni C, Fradkin A, Cian L, Pennycook G (2022) News from generative artificial intelligence is believed less. In: ACM conference on fairness, accountability, and transparency (FAccT), pp 97–106
Maarouf A, Bär D, Geissler D, Feuerriegel S (2023) HQP: a human-annotated dataset for detecting online propaganda. ar**v:2304.14931
Maedche A, Morana S, Schacht S, Werth D, Krumeich J (2016) Advanced user assistance systems. Bus Inf Syst Eng 58:367–370
Maedche A, Legner C, Benlian A, Berger B, Gimpel H, Hess T, Hinz O, Morana S, Söllner M (2019) AI-based digital assistants: opportunities, threats, and research perspectives. Bus Inf Syst Eng 61(4):535–544. https://doi.org/10.1007/s12599-019-00600-8
Matz S, Teeny J, Vaid SS, Harari GM, Cerf M (2023) The potential of generative AI for personalized persuasion at scale. PsyAr**v
Matz SC, Kosinski M, Nave G, Stillwell DJ (2017) Psychological targeting as an effective approach to digital mass persuasion. Proc Natl Acad Sci 114(48):12,714-12,719
Metz C (2023) Instant videos could represent the next leap in A.I. technology. https://www.nytimes.com/2023/04/04/technology/runway-ai-videos.html, accessed 25 Aug 2023
Mirsky Y, Lee W (2021) The creation and detection of deepfakes: a survey. ACM Comput Survey 54(1):1–41
Morana S, Maedche A, Schacht S (2019) Designing process guidance systems. J Assoc Inf Syst pp 499–535, https://doi.org/10.17705/1jais.00542
Ng A, Jordan M (2001) On discriminative vs. generative classifiers: a comparison of logistic regression and naive Bayes. In: Advances in Neural Information Processing Systems, vol 14, pp 841–848, https://papers.nips.cc/paper_files/paper/2001/hash/7b7a53e239400a13bd6be6c91c4f6c4e-Abstract.html, accessed 25 Aug 2023
OpenAI (2022) Introducing ChatGPT. https://openai.com/blog/chatgpt, accessed 25 Aug 2023
OpenAI (2023a) GPT-4 technical report. ar**v:2303.08774
OpenAI (2023b) How should AI systems behave, and who should decide? https://openai.com/blog/how-should-ai-systems-behave, accessed 25 Aug 2023
Park JS, O’Brien JC, Cai CJ, Morris MR, Liang P, Bernstein MS (2023) Generative agents: interactive simulacra of human behavior. ar**v:2304.03442
Peres R, Schreier M, Schweidel D, Sorescu A (2023) On ChatGPT and beyond: how generative artificial intelligence may affect research, teaching, and practice. Int J Res Market 40:269–275
Rabinowitz NC, Perbet F, Song HF, Zhang C, Eslami SMA, Botvinick MM (2018) Machine theory of mind. In: International conference on machine learning, PMLR, vol 80, pp 4215–4224, http://proceedings.mlr.press/v80/rabinowitz18a.html, accessed 25 Aug 2023
Rai A (2020) Explainable AI: from black box to glass box. J Acad Market Sci 48:137–141
Ramaswamy V, Ozcan K (2018) What is co-creation? An interactional creation framework and its implications for value creation. J Bus Res 84:196–205
Reisenbichler M, Reutterer T, Schweidel DA, Dan D (2022) Frontiers: supporting content marketing with natural language generation. Market Sci 41(3):441–452
Rombach R, Blattmann A, Lorenz D, Esser P, Ommer B (2022) High-resolution image synthesis with latent diffusion models. In: IEEE/CVF conference on computer vision and pattern recognition, pp 10684–10695
Sandkuhl K, Fill H, Hoppenbrouwers S, Krogstie J, Matthes F, Opdahl AL, Schwabe G, Uludag Ö, Winter R (2018) From expert discipline to common practice: a vision and research agenda for extending the reach of enterprise modeling. Bus Inf Syst Eng 60(1):69–80. https://doi.org/10.1007/s12599-017-0516-y
Schoormann T, Möller F, Hansen MRP (2021) How do researchers (re-)use design principles: An inductive analysis of cumulative research. In: The Next Wave of Sociotechnical Design, Springer, Cham, Lecture Notes in Computer Science, pp 188–194, https://doi.org/10.1007/978-3-030-82405-1_20
Schoormann T, Stadtländer M, Knackstedt R (2023) Act and reflect: integrating reflection into design thinking. J Manag Inf Syst 40(1):7–37. https://doi.org/10.1080/07421222.2023.2172773
Schoormann T, Strobel G, Möller F, Petrik D, Zschech P (2023) Artificial intelligence for sustainability: a systematic review of information systems literature. Commun AIS 52(1):8
Schramowski P, Turan C, Andersen N, Rothkopf CA, Kersting K (2022) Large pre-trained language models contain human-like biases of what is right and wrong to do. Nat Machine Intell 4(3):258–268
Schwartz R, Dodge J, Smith NA, Etzioni O (2020) Green AI. Commun ACM 63(12):54–63
Schöbel S, Schmitt A, Benner D, Saqr M, Janson A, Leimeister JM (2023) Charting the evolution and future of conversational agents: a research agenda along five waves and new frontiers. Inf Syst Front. https://doi.org/10.1007/s10796-023-10375-9
Senoner J, Netland T, Feuerriegel S (2022) Using explainable artificial intelligence to improve process quality: evidence from semiconductor manufacturing. Manag Sci 68(8):5704–5723
Shin M, Kim J, van Opheusden B, Griffiths TL (2023) Superhuman artificial intelligence can improve human decision-making by increasing novelty. Proc Natl Acad Sci 120(12):e2214840,120
Shollo A, Hopf K, Thiess T, Müller O (2022) Shifting ML value creation mechanisms: a process model of ML value creation. J Strateg Inf Syst 31(3):101,734. https://doi.org/10.1016/j.jsis.2022.101734
Siebers P, Janiesch C, Zschech P (2022) A survey of text representation methods and their genealogy. IEEE Access 10:96,492-96,513. https://doi.org/10.1109/ACCESS.2022.3205719
Silva N, Sousa P, Mira da Silva M (2021) Maintenance of enterprise architecture models. Bus Inf Syst Eng 63(2):157–180. https://doi.org/10.1007/s12599-020-00636-1
Slack D, Krishna S, Lakkaraju H, Singh S (2023) Explaining machine learning models with interactive natural language conversations using TalkToModel. Nat Machine Intell 5:873–883
Smits J, Borghuis T (2022) Generative AI and intellectual property rights. Law and artificial intelligence: regulating AI and applying ai in legal practice. Springer, Heidelberg, pp 323–344
Spitale G, Biller-Andorno N, Germani F (2023) AI model GPT-3 (dis) informs us better than humans. Sci Adv 9:eadh1850
Strobelt H, Webson A, Sanh V, Hoover B, Beyer J, Pfister H, Rush AM (2023) Interactive and visual prompt engineering for ad-hoc task adaptation with large language models. IEEE Transact Visual Comput Graphics 29(1):1146–1156. https://doi.org/10.1109/TVCG.2022.3209479
Susarla A, Thatcher RGJB, Sarker S (2023) Editorial: the janus effect of generative AI: charting the path for responsible conduct of scholarly activities in information systems. Inf Syst Res 34(2):399–408. https://doi.org/10.1287/isre.2023.ed.v34.n2
Sutskever I, Vinyals O, Le QV (2014) Sequence to sequence learning with neural networks. Adv Neural Inf Process Syst 27:3104–3112
Teubner T, Flath CM, Weinhardt C, van der Aalst W, Hinz O (2023) Welcome to the era of ChatGPT. Bus Inf Syst Eng 65(2):95–101. https://doi.org/10.1007/s12599-023-00795-x
Unsal S, Atas H, Albayrak M, Turhan K, Acar AC, Doğan T (2022) Learning functional properties of proteins with language models. Nat Machine Intell 4(3):227–245
van der Aalst WMP, Bichler M, Heinzl A (2018) Robotic process automation. Bus Inf Syst Eng 60(4):269–272. https://doi.org/10.1007/s12599-018-0542-4
Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser Ł, Polosukhin I (2017) Attention is all you need. Adv Neural Inf Process Syst 30:6000–6010
Vernadat F (2020) Enterprise modelling: research review and outlook. Comput Indust 122(103):265. https://doi.org/10.1016/j.compind.2020.103265
Vidgof M, Bachhofner S, Mendling J (2023) Large language models for business process management: opportunities and challenges. In: Business process management forum. Lecture Notes in Computer Science, Springer, Cham, pp 107-123
von Zahn M, Feuerriegel S, Kuehl N (2022) The cost of fairness in AI: evidence from e-commerce. Bus Inf Syst Eng 64:335–348
Wolfe R, Banaji MR, Caliskan A (2022) Evidence for hypodescent in visual semantic AI. In: ACM conference on fairness, accountability, and transparency, pp 1293–1304
Ziegler DM, Stiennon N, Wu J, Brown TB, Radford A, Amodei D, Christiano P, Irving G (2019) Fine-tuning language models from human preferences. ar**v:1909.08593
Zilker S, Weinzierl S, Zschech P, Kraus M, Matzner M (2023) Best of both worlds: combining predictive power with interpretable and explainable results for patient pathway prediction. In: Proceedings of the 31st European Conference on Information Systems (ECIS), Kristiansand, Norway
Zschech P, Horn R, Höschele D, Janiesch C, Heinrich K (2020) Intelligent user assistance for automated data mining method selection. Bus Inf Syst Eng 62(3):227–247. https://doi.org/10.1007/s12599-020-00642-3
Acknowledgements
During the preparation of this Catchword, we contacted all current department editors at BISE to actively seek their feedback on our suggested directions. We gratefully acknowledge their support.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Author information
Authors and Affiliations
Corresponding author
Additional information
Accepted after one revision by Susanne Strahringer.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Feuerriegel, S., Hartmann, J., Janiesch, C. et al. Generative AI. Bus Inf Syst Eng 66, 111–126 (2024). https://doi.org/10.1007/s12599-023-00834-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12599-023-00834-7