Keywords

3.1 Introduction

Digital transformation (DT) and sustainability transformation (ST) are dominant transformational forces. In the past few years, DT has been driven by rapid advancements in digital technologies and has had profound impacts on individuals, organisations, and society (e.g., Vial, 2019; Wessel et al., 2021). Emerging digital technologies, such as digital platforms and Artificial Intelligence (AI), are advancing the ability to collect and process ever-larger volumes of data, make predictions based on that data, and generate solutions. Current DT research mainly focuses on such technological progress changing value creation paths and related positive and negative impacts on different levels of analysis (Hanelt et al., 2021; Vial, 2019). At the same time, concerns about environmental degradation, social inequality, as well as economic instability shape market dynamics and have accelerated discussions about digital sustainability and digital resilience (Boh et al., 2023; Kotlarsky et al., 2023). ST depends on the vital role of digital technologies such as AI-enabled systems in addressing environmental and societal challenges to facilitate the development of innovative solutions and systemic changes (Lehnhoff et al., 2021; Watson et al., 2010). AI-enabled monitoring and analysis of data like CO2 levels and forecast data of extreme weather events play major roles in environmental concerns of ST.

Given the need to pursue both key transformations simultaneously, some businesses and regulators (e.g., European Commission, 2022) have identified a synergistic relationship between DT and ST. Businesses that use an integrated approach to deal with both transformations at once appear to be more successful than those that focus on one at a time (Ollagnier et al., 2021). The European Commission (2022) identifies several applications for an integrated DT and ST approach, including systematic management of supply chains and financial flows, developing monitoring frameworks that measure well-being beyond economic goals, and advancing secure data-sharing frameworks.

Despite these potential synergies, the academic discourse on DT and ST has evolved in relative isolation. Only recently information systems (IS) research started to discuss the potential of an integrated transformation of DT and ST, using the label Twin Transformation (e.g., Christmann et al., 2024; Graf-Drasch et al., 2023). Christmann et al., (2024, p. 7) characterise Twin Transformation as “a value-adding interplay between digital and sustainability transformation efforts that improve an organisation by leveraging digital technologies for enabling sustainability and leveraging sustainability for guiding digital progress.” Thus, Twin Transformation leverages DT to develop digital solutions that improve ST to provide the goals and insights that are required to design those digital solutions.

In this chapter, we argue that IS researchers and practitioners can play a role in further integrating DT and ST to capitalise on their synergistic potential, acknowledging that IS are embedded in larger systems where human action affects and is affected by the natural environment (Christmann et al., 2024). Specifically, we highlight how AI—the ever-evolving frontier of computational advancement (Berente et al., 2021)—will play a pivotal role in realising Twin Transformation. We develop a framework for AI-enabled Twin Transformation to show how AI-enabled systems can help to overcome the boundaries of human rationality in addressing the complex problem space that exists at the intersection of DT and ST.

The remainder of this chapter is structured as follows. Section 3.2 outlines the Twin Transformation concept and highlights how the problem spaces of DT and ST overlap. Section 3.3 describes the role of AI-enabled systems in contributing to DT’s and ST’s joint solution space. Finally, Sect. 3.4 concludes the chapter with a discussion of our framework’s implications for IS research and practice.

3.2 Twin Transformation: Converging the Problem Spaces of Digital Transformation and Sustainability Transformation

Twin Transformation integrates DT’s and ST’s problem spaces providing a joint solution space at their interface. These problem spaces comprise the respective challenges of the individual transformations, while they are addressed in an integrated manner in the Twin Transformation solution space. Such integration may appear contradictory at first, as DT initiatives typically focus on economic concerns (e.g., efficiency improvement, sales increase) (Vial, 2019), whereas ST initiatives are motivated by social and environmental concerns (Schoormann, 2020; Seidel et al., 2013).

The DT problem space refers to digital innovations that transform aspects of private and professional lives, organisations’ value propositions (Wessel et al., 2021), and society’s interconnectedness (Mousavi Baygi et al., 2021). At the individual level, digital technologies redefine communication, collaboration, workplace design, and work practices (sometimes referred to as the future of work). At the organisation level, DT affects processes, products, services, and business models (Vial, 2019). At the societal level, an interconnected techno-society unfolds in which digital technologies create and shape reality instead of only representing it (Baskerville et al., 2019). At all levels, DT involves continuous change and causes significant tensions between the ‘old’ and the ‘new’ (Drechsler et al., 2020), requiring flexibility and acceptance of a new culture (Svahn et al., 2017). As a result, the success of DT is often only partial—but the partial success is also because its complex drivers and effects are still poorly understood (Gurbaxani & Dunkle, 2019).

The ST problem space refers to social, environmental, and economic sustainability issues related to individuals, organisations, and society. Individuals can have a positive impact on sustainability by making sustainable consumption choices, while organisations can contribute by empowering individuals to make sustainable consumption choices and to use their power to improve global sustainability. The effect of organisational behaviour should not be underestimated, as, for example, the energy sector in the European Union (EU) is responsible for two thirds of the greenhouse gas (GHG) emissions of the EU (European Parliament, 2023). At the societal level, legislators use regulations to steer individuals’ and organisations’ behaviour and support intergenerational justice by mitigating biodiversity losses and natural disasters to ensure that future generations can continue to live in a world worth living in (Ekardt et al., 2023). Overall, ST uses the underlying mechanisms and links among the three levels of sustainability to shape and guide its means and ends.

Building on insights from IS research on DT and ST problem spaces, recent publications focus on the intersection where solutions address DT- and ST-related problems simultaneously. Zimmer and Järveläinen (2022), for instance, apply the triple-bottom line of economic, environmental, and social sustainability to DT research and provide a framework for sustainable and digital co-transformations. Graf-Drasch et al. (2023) analyse Twin Transformation on various organisational levels using an integrative work system perspective to describe the interplay of DT and ST and guide organisations in their Twin Transformation. Christmann et al. (2024) examine dynamic capabilities of making DT sustainable and enabling the digitalisation of ST processes to realise Twin Transformation. In this context, particularly because of their learning abilities, AI-enabled systems are recognised as the current technological frontier for developing dynamic capabilities in transformational DT and ST, and the specific role of AI in Twin Transformations warrants our attention.

3.3 A Framework for AI-enabled Twin Transformation

Twin Transformation is rooted in two complex and overlapping problem spaces, each rife with multidimensional problems that are too complex and too large for humans to navigate. DT and digital technologies like AI-enabled systems open many opportunities to address the multi-layered challenges of sustainability, which are often characterised by uncertain interdependencies and nonlinearities (Malhotra et al., 2013; Schoormann, 2020; Watson et al., 2010;). The complexity that results from DT’s almost infinite opportunities and ST’s multidimensional dependencies make it difficult for humans to evaluate the value of a (digital) solution design (Rai, 2017), so Twin Transformation is a prime example of problems that require application of AI-enabled solutions’ predictive and generative capabilities to overcome the boundaries of human rationality (Berente et al., 2021). Through their capacity to learn, make predictions, support decision-making, and generate new solutions, AI can help to build socio-technical systems that have the requisite variety (Ashby, 1991) needed to address complex economic, environmental, and social concerns simultaneously.

The interplay between DT and ST is enabled by networks of sensitised objects, which generate the data streams that provide fodder for AI-enabled systems. AI-enabled systems can process large amounts of data that form the basis for their ability to learn (i.e. improve through data and experience) and to be autonomous (i.e. having the ability to act without human intervention) in an expanding range of contexts (Agrawal et al., 2018; Berente et al., 2021). Moreover, AI-enabled systems can provide predictions and generate design options that can inform design decisions and lead to new data streams. They can find patterns in large amounts of unstructured data and generate novel artefacts (e.g., through generative AI), thus helping to clarify phenomena related to sustainability and informing appropriate design interventions (Padmanabhan et al., 2022). ST requires AI-enabled systems to learn about a transformation’s consequences, such as the gains that are likely from implementing aspects of the Circular Economy (Zeiss et al., 2021).

In the AI-enabled Twin Transformation solution space, AI-enabled systems facilitate identification of patterns and structuring of pertinent data (streams), thereby catalysing Twin Transformation efforts (Christmann et al., 2024). AI-enabled solutions for Twin Transformation learn from incoming data streams from DT, while the ST aspect is reflected in providing goals and occasions for generating that data, thereby guiding the design of new solutions (Graf-Drasch et al., 2023). Figure 3.1 captures the dual dynamics that underlie AI-enabled Twin Transformation, including the role of data streams and AI-enabled systems.

Fig. 3.1
A Venn diagram. The left and right circles are labeled the digital transformation problem space and the sustainability transformation problem space. The overlapped area of both cirles is labeled A I enabled twin transformation solution space at the top anddata streams and A I enabled system at the bottom.

Framework for AI-enabled Twin Transformation

Full size image

We conceive of the AI-enabled Twin Transformation solution space as being realised through AI-enabled solutions at the individual, organisational, and societal levels. AI-enabled Twin Transformation solutions are based on the capabilities of AI-enabled systems, while their design is guided by sustainability principles or purpose. Table 3.1 highlights examples of such AI-enabled solutions for Twin Transformation at these three levels of analysis. The examples show that DT and ST interact synergistically, which results in contributions to sustainability objectives as well as a positive impact on digitalisation.

Table 3.1 Examples of AI-enabled solutions for twin transformation
Full size table

Recognising that AI-enabled Twin Transformation is a boundary-spanning, holistic transformation, questions for research, and practice arise at the three levels of analysis (Fig. 3.2). First, individual-level behaviour represents the basis for change on all other levels. Individual-level Twin Transformation involves both leveraging data streams and AI to learn about individual behaviour’s impacts on sustainability, and designing digital applications to guide individuals towards sustainability-oriented behaviour (Bashir, 2022) while ensuring technology acceptance (Venkatesh et al., 2016). Organisation-level research and practice should use AI-enabled systems to explore pattern identification and the impact of organisational activities on sustainability to support the design of cost and resource-efficient digital processes, products, services, and business models (El Hilali et al., 2020). Societal-level Twin Transformation integrates DT’s impact on sustainability and ST’s impact on digitalisation to influence regulations that measure and reward integrated DT and ST initiatives (European Commission, 2022).

Fig. 3.2
An onion diagram represents the key questions about A I artificial intelligence in 3 layers. The 3 circles for the 3 layers are labeled individual behaviour, organisational change, and societal regulation. Each of the layers includes 2 different questions.

Key questions about AI-enabled twin transformation on three levels of analysis

Full size image

3.4 Implications for Information Systems Research and Practice

To foster leadership and develop mitigation strategies related to the current challenges for DT and ST, such as how to motivate individuals to use new digital technologies or how to enable organisations to measure their impact on climate change, IS researchers and practitioners should focus on AI-enabled Twin Transformation. We identify three implications of such a focus that emerge from this view.

First, Twin Transformation that builds on AI-enabled systems and data streams requires capitalising on the learning and designing cycles simultaneously. Predictions facilitate better designs that can produce new streams of economic, environmental, and social data. Bringing together DT and ST perspectives can result in a virtuous cycle of learning and design activities. Not every IS study has to do both, but we suggest that they at least build on each other cumulatively. Twin Transformation is complex, and complexity can be dealt with through decomposition (Baldwin & Clark, 2000; Simon, 1996). For instance, learning that a particular digital component achieves a particular goal in a particular system (e.g., sensors that monitor the operation of production processes) can provide the foundation for further, more complex designs that produce more complex data streams (e.g., for assessing and certifying the GHG emissions generated in the supply chain). Managing AI-enabled systems in Twin Transformation requires managing the learning and designing cycles that alternate or blend.

Second, Twin Transformation research integrates DT and ST problem spaces, thus opening a new solution space at their intersection, where AI-enabled systems catalyse Twin Transformation solutions that learn from DT to foster sustainability and exploit ST’s guidance for DT design (Christmann et al., 2024; Graf-Drasch et al., 2023). However, using AI-enabled systems can be resource-intensive (e.g., energy consumption) and subject to social biases (e.g., gender bias), thus negatively affecting environmental and social sustainability. Hence, practitioners and researchers must account for address, and improve the sustainability of AI-enabled systems across their entire lifecycle to exploit all of Twin Transformation’s potential (van Wynsberghe, 2021).

Third, our research offers an outlook on the future of AI-enabled systems and Twin Transformation’s interplay in practice. Individuals, organisations, and society deal with the infinite possibilities of AI-enabled solutions. Our framework supports individuals, organisations, and society in connecting AI-enabled solutions and the objectives of Twin Transformation to leverage digital and sustainable advantages. By highlighting the role of data streams and AI-enabled systems in Twin Transformation, our work presents practitioners with a fresh strategic perspective on integrating DT and ST problem spaces.

In conclusion, we argue that Twin Transformation is the pivotal transformation for this and the coming decades. Joint discourse grounded in research on AI-enabled systems, IS for environmental sustainability (i.e., Green IS, Green IT), and DT can help to clarify the relationship between the two transformations, namely digital and sustainability transformation, and explorations of the AI-enabled Twin Transformation solution space to unearth digital and sustainable results.