Content uploaded by Timo Böttcher
Author content
All content in this area was uploaded by Timo Böttcher on May 03, 2022
Content may be subject to copyright.
Value Drivers of Artificial Intelligence
Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022
1
Value Drivers of Artificial Intelligence
Completed Research
Timo Phillip Böttcher
Technical University of Munich
timo.boettcher@tum.de
Michael Weber
Technical University of Munich
mic.weber@tum.de
Jörg Weking
Technical University of Munich
joerg.weking@tum.de
Andreas Hein
Technical University of Munich
andreas.hein@tum.de
Helmut Krcmar
Technical University of Munich
helmut.krcmar@tum.de
Abstract
Artificial intelligence (AI) holds great potential for firms to create new business models and gain
competitive advantages. While some pioneers are effectively leveraging AI, most firms are struggling to
capitalize on the opportunities for value creation. Previous research has highlighted the performance
benefits, success factors, and challenges of adopting AI. However, the value drivers of AI, specifically
regarding how AI creates value, remain unclear and need exploration so that firms can adapt their value
creation to leverage the potential. To clarify how AI creates value, we conduct a case survey of 61 firms to
identify six value drivers: efficiency, novelty, knowledge from data, ecosystem, personalization, and human
resemblance. We discuss how these value drivers differ from other digital technologies. For practitioners,
we provide valuable insights into the business value of AI and business model (BM) design opportunities to
build on.
Keywords
Artificial intelligence, value creation, business model, case survey.
Introduction
Artificial intelligence (AI) is a highly promising technology that influences the value creation process of
firms (Davenport and Ronanki 2018). By 2025, the business value created globally by AI is expected to
nearly double from today’s three trillion US$ to over five trillion US$ (Gartner 2019). Some firms, like Uber,
have recognized the opportunities of AI early on and pioneer its application; Uber uses AI for fraud
detection, risk assessment, marketing budget allocation, driver and rider matching, route optimization, and
driver onboarding. Such firms have successfully aligned their business models (BMs) with AI and created a
competitive advantage (Lee et al. 2019). Simultaneously, many firms have hardly implemented any AI
applications yet (Ransbotham et al. 2017). Due to the complexity involved, these firms often face significant
issues at all levels, such as a missing AI strategy, a lack of technical expertise, and an ill-defined
understanding of how AI can create value (Lee et al. 2019).
To fully leverage AI and create value, firms need to embed it in a BM (Teece 2010). Firms need to adapt
their BMs to take full advantage of AI’s benefits, but also need AI and other digital innovations to create a
competitive digital BM (Chesbrough 2007). A BM is a tool for the development, innovation, and evaluation
of firms’ business logic (Veit et al. 2014). It is “the design or architecture of how value is generated,
delivered, and monetized” (Teece 2010). If AI is included, firms need to understand how it can create value.
However, advances in AI and other deep-tech technologies have led to increased complexity as they increase
interdependencies between technologies, processes, and firms in the ecosystem (Benbya et al. 2020).
Value Drivers of Artificial Intelligence
Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022
2
Research has analyzed AI from multiple perspectives, such as financial and strategic performance of firms
(Reis et al. 2020), adoption and barriers to adopting (Alsheibani et al. 2020; Jöhnk et al. 2021), BM
characteristics (Weber et al. 2021), AI-capabilities (Schmidt et al. 2020), and top management influence
(Li et al. 2021).
AI “as the frontier of computational advancements that references human intelligence in addressing ever
more complex decision-making problems” (Berente et al. 2021) creates manifold opportunities for value
creation, such as autonomous robots organizing warehouses (e.g., Nimble Robotics) and intelligent
software optimizing sales (e.g., Outreach), time management (e.g., Clockwise), or product development
(e.g., BenchSci). Yet, the value drivers of AI, describing how AI creates value, remain unclear and need
clarity so that firms can adapt their value creation to leverage AI’s potential (Schmidt et al. 2020).
Various theoretical and practical models exist in the literature to assess technology-driven value creation,
such as the “sources of value creation in e-business” model (Amit and Zott (2001). This model consists of
four value drivers in BMs enabled by the Internet: efficiency, novelty, lock-in, and complementarities (Amit
and Zott 2001). The Internet digitalized value creation by enabling online transactions, such as
matchmaking on digital platforms, which was done offline before. AI changes value creation by enabling
new processes, products, and BMs, that were not feasible before (Borges et al. 2021). Therefore, we propose
the following research question to analyze how AI creates value and how it differs from previous digital
innovations like the Internet: what are the sources of value creation in AI-enabled business models?
We conduct a case survey (Larsson 1993) to reveal sources of value creation in AI-enabled BMs. We
developed a structured questionnaire based on extant literature on value creation, digital BMs, and
applications of AI. We identified 61 firms with AI-enabled BMs and answered the questionnaire for each
firm using publicly available information. For the questionnaire analysis, we conducted open, axial, and
selective coding (Corbin and Strauss 1990) to identify value drivers of AI and create a model of how AI
creates value in BMs. The model displays the variety of AI value creation and how it differs from previous
digital innovations. For practice, we provide insights into the business value of AI and BM design options
firms can build upon.
Related Work
A BM represents a firm’s business logic (Veit et al. 2014). It defines how a firm ”creates and delivers value
to customers, and then converts payments received to profits” (Teece 2010). In their BM Canvas,
Osterwalder and Pigneur (2010) formulate nine elements. These are the value proposition, customer
segments, customer relationships, sales channels, revenue streams, key partners, key activities, key
resources, and cost structure of the BM.
Along digital transformation, the BM has become relevant in articulating the value creation from digital
innovations. Vial (2019) even describes the BM as a central outcome of the digital transformation, that is,
a digitally transformed BM or just digital BM. A BM is digital if digital innovations bring fundamental
changes in the way business is done, transacted, and value is created (Veit et al. 2014). Thus, many digital
BMs have fundamentally changed the way value is created (Steininger 2019). For example, a very prominent
type of digital BMs is digital platforms, such as Uber or Amazon, creating multi-sided markets by matching
multiple actors via the Internet.
Value creation is a core component of BMs and summarizes all activities related to creating the value
proposition for the customer (Amit and Zott 2001). The topic of how digital innovations enable new forms
of value creation has been a major focus of BM research in information systems (Steininger 2019). It has
become indispensable in the business world and influences all core processes and operations in firms (Wang
et al. 2014). Digital technologies are resources, through which value can be created (Lee et al. 2014) and
(Steininger 2019) articulated four roles of digital technology in value creation to enable digital BMs: a
facilitator of business operations (e.g., robot process automation for business processes), a mediator of
value delivery to the customer (e.g., online shops), an outcome of the value creation (e.g., software
development), and a ubiquity where the entire BM is digital (e.g., digital platforms). Already with the
emergence of the Internet Amit and Zott (2001) examined how it enabled new forms of value creation. They
identified four value drivers: first, value is created on the Internet through more efficient transactions
(“efficiency”); second, due to a complementary product offering, products are often sold together, which in
turn increases the value created (“complementarities”); third, fewer customers switch to competitors due
Value Drivers of Artificial Intelligence
Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022
3
to the “lock-in” effect; and fourth, the Internet enables completely new types of BMs (“novelty”) (Amit and
Zott 2001).
Following Berente et al. (2021), AI is “the frontier of computational advancements that references human
intelligence in addressing ever more complex decision-making problems,” such as predicting the shape of
proteins from the sequence of its amino acids. Thus, it represents what Information Technology is currently
capable of and the opportunities for value creation seem endless. To understand how AI can create value
today, we summarize three major applications to demonstrate its capabilities as a comprehensive list is
beyond this paper’s scope.
First, AI can analyze large data sets and discover patterns and logical relationships. Thereby, it can make
predictions concerning time series or the probability of events (Agrawal et al. 2019). Compared to
traditional data analysis, AI models are constantly evolving through self-learning algorithms, implying that
the ability to make predictions or categorize things is constantly improving (Davenport and Ronanki 2018).
Second, AI can recognize, classify, and process objects, images, videos, and text. Machine vision enables AI
systems to perceive visual content such as images and videos in their environment just like humans
(Goodfellow et al. 2016). This property is important for several visual recognition AI applications. Natural
Language Processing (NLP) enables the understanding of human language by AI systems (Goodfellow et al.
2016). NLP can recognize languages, analyze grammar and sentence structures, and understand the
meaning of written and spoken language. For example, AI can recognize spam, fake news, and even
emotions. Third, AI can autonomously control physical objects, such as robots (Murphy 2019). AI responds
to changes in the environment and controls objects accordingly. For example, it enables an autonomously
driving car to handle different traffic situations, known and unknown.
Ever since AI became an economically relevant technology, research looked into the competitive benefits of
its implementation and how it provides benefits for firms, their strategies, and BMs (Brynjolfsson and
Mcafee 2017; Teece 2018). AI and its related technologies, such as machine learning, are already shown to
positively influence firm performance (Reis et al. 2020; Wamba-Taguimdje et al. 2020). BM research has
developed taxonomies of AI-enabled BMs (Weber et al. 2021) and AI-enabled value propositions in
particular (Nowacki 2019) to analyze how firms build viable BMs based on AI. The value creation aspect
has been analyzed from detangled viewpoints, such as multi-stakeholder perspective (Güngör 2020), value
co-creation (Kaartemo and Helkkula 2018), or AI-capabilities (Schmidt et al. 2020). Borges et al. (2021)
propose a conceptual framework connecting AI technologies and business strategy and suggest four sources
of value creation based on extant literature that needs further investigation.
However, firms still face various organizational challenges, such as strategic goals, business value
demonstration, or top management support, when implementing AI that hinder successful value creation
(Alsheibani et al. 2020; Someh et al. 2020). In summary, extant research demonstrates the benefits and
challenges of AI. However, more research is needed to explain how and why AI creates value and if the value
drivers differ from the ones known from other digital innovations (Amit and Zott 2001; Benbya et al. 2020;
Borges et al. 2021; Teece 2018).
Methodology
To identify the value drivers of AI, we use the case survey method to create generalizable insights from
qualitative data (Larsson 1993). The case survey allows us to analyze shared elements of AI value creation
and develop a theoretical model of their interdependencies. The method consists of four steps. First, we
developed a structured questionnaire based on extant literature. Second, we identified our case sample of
firms with AI-enabled BMs. Third, we answered our questionnaire for all firms using public information.
Fourth, the qualitative data in the questionnaire were analyzed using open, axial, and selective coding
(Corbin and Strauss 1990) to develop the value drivers and a model of AI value creation.
The questionnaire consists of four parts. The first part comprises general information about the firms, such
as industry, headquarter location, and initial public offering (IPO) date. The second part uses the BM
Canvas (Osterwalder and Pigneur 2010) to capture all components of the BM including how AI is relevant
for the value proposition, the revenue model, or the customer relationships. The third part examines what
the firms specifically use AI and data for. This includes a summary of specific AI use cases, a description of
the role of data and what it is used for, and if the AI affects products, processes, or services. The fourth part
Value Drivers of Artificial Intelligence
Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022
4
records how Amit and Zott’s (2001) Amit and Zott (2001)four value drivers apply to the analyzed AI-
enabled BMs.
We created the case sample for our case survey from Crunchbase. We selected firms assigned to the
"Artificial Intelligence” industry. Following Amit and Zott (2001), we only included firms listed on a stock
exchange to ensure that sufficient and reliable information was available. This resulted in 155 relevant firms
that were analyzed if they fit our research purpose. Therefore, we applied exclusion criteria (Larsson 1993):
firms that are no longer active (-4), firms without information in English (-42), firms without sufficient
public information to answer our questionnaire (-32), and firms that we did not consider to be
implementing an AI-enabled BM (-16; e.g., WISeKey provides a digital identity platform that uses AI
elements, but does not rely on it). The final sample had 61 relevant firms.
Figure 1. Coding procedure
We collected information about the firms' AI usage from several data sources (Amshoff et al. 2015), such as
the firms’ publications, such as their financial reports (e.g., 10-K forms), firm announcements, and
websites. Since we restricted our case sample to publicly listed firms, we assumed the correctness of these
information. We triangulated this information with other sources such as renowned newspapers’ (e.g.,
Financial Times, Wall Street Journal) articles, academic and practitioner case studies, analysts’ investment
reports, and public databases to get a comprehensive view on the firms’ strategic use of AI and avoid
potential biases (Amshoff et al. 2015; Yin 2017). In total, we collected 341 sources for the completion of the
questionnaire.
Figure 1 shows how abstract AI topics were initially derived from the individual results of the case survey.
We first collected text fragments as open codes. With axial coding, these were grouped into different,
abstracted AI topics, such as “self-learning.” Finally, we applied selective coding to derive themes
Open coding Selective codingAxial coding
AI value drivers
Results from the case survey Abstract AI topics
New products
New business models
Novelty
n = 61
Together with other
technologies
Partnerships Ecosystem
n = 41
Value co-creation
Personaliza-
tion
n = 23
Self-learning
Intuition and
behavior like a human
Human
resemblance
n = 38
Personalization
•Plus (autonomous dr iving): Enables co mpletely new business models for carmakers and mobi lity
providers.
•Alfi (ad vertising pla tform): AI-based personalized adve rtising based on facial features.
•Zymergen (biotechnology): AI helps find patterns i n genomic data, which are then used to develop
new products
•Treatment.com (medical diagnosis): AI-based medical diagnosis on one platform
•Darktrace ( information securi ty): Software l earns independent ly from past cyber attacks and
continues to evolve
•KPIT Technol ogies (autonomou s driving): Improving autonomous driving throu gh self-learning
algorithms.
•Bigtincan (sales and learning platform): AI is used to personalize learning content for individual
users.
•Latern Pharma (pharmaceuti cal manufacturer): Person alized cancer therapy through the use of AI.
•Recursion (b iotechnology): AI i s used to develop new medicines in collaboration with customers.
•Veritone (d evelopment platform): p latform on which customers can easily develop new AI
applications
•TuSimple (autonomous driv ing): strategic partner ships needed with truck manufact urers and other
suppliers
•Intel (semiconductor manufacturer): AI Builders program links different AI companies
•Tuya (IoT platform): AI enables voice recognition and facial recognition to be built into IoT devices.
•Windfal l Geotek (geotechnology): Using drones and AI to identify landmines
•Plus (autonomous dr iving): Recogniti on of other road users. Road structures , as well as control of
the vehicle.
•Artifici al Solutions ( customer dialog systems): Chatbo ts and voice a ssistants are used for
communication.
•mCloud (building management): AI is used to optimize the energy consumption of buildings and
facilities.
•Yext (knowledge management): Optimization of search queries on company websites using AI.
Optimization
Automation
Efficiency
n = 61
•Upstart (bank): With the help of AI, c redit checks and credit decisions are fully automate d
•Lemonade (online insurance): AI autonomously makes decisions on claims settlement and contract
offers
Predictions
Data Analysis
Knowledge
from data
n = 52
•Bioxcel (biotechnology): analysis of data to find mechanisms and causes of diseases
•Mirriad (video analysis): Identifying possible advertising slots in films
•Appier (marketing platform): AI makes it possible to predict customer behavior
•Neural Pok et (data analytics): AI is used to predict fashion trends
Value Drivers of Artificial Intelligence
Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022
5
representing the value drivers of AI. We further analyzed the relationships between the value drivers,
looking into dependencies and interactions, to create a theoretical model of AI value creation.
Results
Case Sample
Our case sample consists of 61 firms. The firms are software firms (56%, n = 34), platform firms (25%, n =
15), biotechnology firms (10%, n = 6), autonomous driving firms (3%, n = 2), computer chip firms (3%, n =
2), and finance firms (3%, n = 2). Most are headquartered in North America (56%, n = 34) followed by Asia
(25%, n = 15) and Europe (19%, n = 12). For the date of foundation, 48% (n = 30) were founded in the last
ten years, and 30% (n = 18) from 2001 to 2010. This temporal trend is even more apparent in the date of
the IPO: 78% (n = 48) of the firms had their IPO in the last ten years.
Value Drivers
Analyzing the 61 firms revealed that value creation through AI can be mapped by the six inductively derived
value drivers of efficiency, novelty, knowledge from data, ecosystem, personalization, and human
resemblance. Furthermore, we identified strong interdependencies between them.
Efficiency. AI contributes to an increase in efficiency; it is primarily used to automate processes and
services, whereby automation here refers to replacing manual human work. An example is Upstart, which
uses AI to automate the entire loan origination process. By using AI, the customer's creditworthiness can
be predicted more accurately. This, in turn, allows the cost of the loan to be cheaper than conventional
loans.
Novelty. A range of novel products, processes, services, and BMs are enabled that would not be possible
without AI. One example is Minerva Intelligence, which has AI find the optimal location for mines. The
traditional mine search process is expensive because it involves a lot of drilling and analysis. Using AI,
Minerva creates 3D models of potential mine sites and predicts the best potential sites.
Knowledge from data. AI is often used to analyze vast amounts of data, gain new insights, and make
predictions. For example, Neural Pocket uses AI to analyze fashion trends by using Deep Learning to
analyze data from social networks to predict new trends. This enables firms to react to new fashion trends
early.
Ecosystem. Additional value can be created through collaborations with partners, customers, and other
stakeholders. For some firms, this ecosystem is even a prerequisite for their business activities. The value
creation takes place within the firm itself and in collaboration with its customers. One example is Veritone,
which offers a platform for its customers to develop AI applications using a modular system. Veritone also
supports its customers in developing AI applications on the platform, meaning that value creation occurs
jointly.
Personalization. Many firms use AI to tailor content, advertising, and even medical therapies to individuals
or groups. The sales and learning platform, Bigtincan, uses AI to provide each user with individualized
learning content and training. As a result, each sales employee receives only the relevant information and
can thus focus more effectively on the critical tasks. AI is needed to determine the individual needs of each
user from the user and behavioral data.
Human resemblance. AI enables bots, robots, and software to behave similarly to people. This human-like
behavior can allow AI to take over complex tasks previously performed by humans. For example, Artificial
Solutions enable customer service that is very similar to interacting with a human representative through
chatbots and intelligent virtual voice assistants.
Dependencies Between the Value Drivers
The analysis of relationships between the six value drivers resulted in Figure 2. Efficiency, novelty,
knowledge from data, personalization, and human resemblance all interact with each other, while
ecosystem has an overarching role in value creation in digital contexts.
Value Drivers of Artificial Intelligence
Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022
6
The relationship between novelty and efficiency described by Amit and Zott (2001) persists regarding AI.
The mechanisms that enable efficiency cannot be realized without the novelty created by new AI-based
products or services. Simultaneously, the efficiency enabled by AI increases the rate of innovation, thus
creating novelty. BioXcel Therapeutics, for example, uses AI to analyze structured and unstructured data
from science, like trial data and scientific publications, to develop new therapies and drugs faster and more
efficiently. Novelty is similarly linked to personalization, knowledge from data, and human resemblance.
For example, ALFI's facial recognition enables personalized advertising that creates novel value for
advertisers. Autonomous driving features from Plus creates new BMs by imitating human drivers.
Knowledge from data enables novelty, for example, by finding patterns in data to develop new products.
Figure 2. Value creation from AI
Efficiency influences human resemblance, especially when AI is used to automate processes that go beyond
rule-based executions and typically require human intuition. For example, with Lemonade, an online
insurance firm, AI evaluates insurance claim data and images to approve or deny claims. This way, AI
accelerates customer service by mimicking human decision-making and increasing efficiency. The link
between the efficiency and knowledge from data value drivers stems from the observation that analyzing
data enables predictions that make processes more efficient. Efficiency and personalization are linked, as
personalization allows direct targeting. For example, AI is used to play ads on Kuasishou Technology's video
platform based on personalized user behavior. This optimizes the placement of advertising for efficient
targeting.
Furthermore, human resemblance and personalization share dependencies for value creation. The human-
like behavior, Chatbots for example, and self-learning capabilities of AI support and improve
personalization. For example, Treatment.com creates an artificial conversation with patients to diagnose
illnesses and recommend personalized treatments. Human resemblance and knowledge from data are
related, especially in the self-learning capability of AI. If an AI creates knowledge from data, it remembers
this knowledge to improve functionality and thereby resembles human capabilities. For example, Darktrace
provides a self-learning AI to detect cyber attacks, that reports suspicious behavior in a system to prevent
cyber attacks and improves its accuracy with every new suspicious event.
Finally, knowledge from data and personalization are related since data analysis is needed to personalize
content or services. As such, knowledge from data is a prerequisite for personalization. Lantern Pharma
illustrates this when using AI for the personalization of cancer therapies based on the knowledge identified
from individual health records and genome data.
Discussion
Most firms struggle to harness the potential of AI and effectively create value. Extant research has identified
the potentials, performance implications, and also challenges to overcome when implementing AI in BMs.
The progress in AI has led to increased complexity for value creation, but also great new opportunities
(Benbya et al. 2020). Yet, the value drivers of AI remained unclear (Schmidt et al. 2020). Thus, firms still
struggle to determine the value that is created from using AI and integrating it successfully in their BMs
Knowledge from data
Personalization
Human resemblance
Efficiency
Ecosystem
Novelty
Value Drivers of Artificial Intelligence
Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022
7
(Amit and Zott 2012; Ransbotham et al. 2017). To understand how AI creates value, and how this is different
from the value creation from other digital innovations (Benbya et al. 2020; Teece 2018), we analyzed 61
firms employing an AI-based BM and identified six value drivers of AI. Based on these value drivers, we
created a model of AI value creation (Figure 2).
Our model shows that value creation from AI is different from previous digital innovations. Similar to the
Internet, AI still creates value through novelty and efficiency, but differ in detail. The Internet mostly
created novel and more efficient ways to perform transactions, such as matchmaking on digital platforms.
While digital platforms and online shops were new BMs, they are “only” digitalized versions of known BMs.
Novelty and efficiency enabled by AI, create new value by enabling processes, products, and BMs that were
not feasible before (Borges et al. 2021; Weber et al. 2021). AI enables value creation by making business
processes more efficient. It allows replacing human labor with automated and intelligent solutions that can
replace human actions (Lee et al. 2019; Raisch and Krakowski 2021; Schmidt et al. 2020). The processes
can be executed faster and with fewer errors through AI, leading to cost and time savings (Wamba-
Taguimdje et al. 2020). The value creation is revealed in the achieved saving, but also in the additional value
that the human workers can create, for example in creative tasks, instead of engaging in the automatable
tasks (Agrawal et al. 2019; West 2018). AI also creates value by enabling novel processes, products, services,
and BMs that are sources of value creation through innovations such as AI (Chesbrough et al. 2018;
Schumpeter 1934). Since AI presents a moving “frontier of computational advancements” (Berente et al.
2021) the value creation through novelty is likely to increase further. The new possibilities offered by the
digital innovations in AI continue to enable new products, services, and BMs such as humanoid robots or
autonomous driving.
Additionally, we identified human resemblance, knowledge from data, personalization, and ecosystem to
complete our model. In the literature, AI is often referred to as human-like (Davenport and Ronanki 2018;
Teece 2018). While AI is still not able to fully mimic humans, it can solve specific problems better than
humans, for example by finding new solutions to a problem, thereby creating new value (Goodfellow et al.
2016; Hartmann 2019). As AI technologies are further improving to mimic human intelligence (Goodfellow
et al. 2016; Wodecki 2019), the opportunities for value creation through human resemblance and the
associated novelty will further increase. Knowledge from data is a core driver of AI value creation (George
and Lin 2017; Hartmann 2019). The ability to analyze vast amounts of data, detect patterns, and predict
future values enables tremendous value creation (Kitsios and Kamariotou 2021). This knowledge creation
goes beyond what humans or traditional computational methods could perform. Creating this new
knowledge presents a competitive advantage for firms (Grant 1996). However, this also requires large
amounts of data for analysis. For AI value creation, data becomes a strategic resource that has to be
collected, managed, and leveraged (Hartmann and Henkel 2020; Mikalef and Gupta 2021). The value
creation from personalization stems from AI’s higher precision in targeting and forecasting individual
needs (Lee et al. 2019). Personalization is possible without AI, but AI allows responses to individual
preferences and habits, emotions recognized in images or text, or complex data combinations such as
genome data in medicine (Nowacki 2019; Wodecki 2019). Finally, the ecosystem takes a special role in value
creation. In many cases, value is no longer created at the firm alone, but together with others. This so-called
value co-creation involves other organizations in the value creation process, for example, creating new
products together with customers (Hein et al. 2019). The ecosystem also becomes relevant for the strategic
resource data. Through collaborations, data can be combined to improve AI application and enhance value
creation (Hartmann and Henkel 2020).
The various potential applications of AI create complexity in its value creation. The growing dependencies
in the ecosystem between technologies, processes, and organizations are even increasing this complexity
(Benbya et al., 2020). It is also reflected in the dependencies between the value drivers for AI value creation.
The value drivers display strong interdependencies to create a value and therefore cannot be regarded in
isolation. The AI-enabled BMs in our case sample always combine multiple value drivers. To create value
from AI, firms need to understand the individual value drivers and their interdependencies. Data and the
creation of knowledge from data is a key driver of AI value creation, which was present for almost all our
cases. Without AI creating new knowledge that could not be created before, value creation will be impaired.
Similarly, the novelty of AI will enable new forms of value creation if technology keeps improving as fast.
For firms, this implies insecurity and complexity for their value creation that needs to keep up.
Value Drivers of Artificial Intelligence
Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022
8
Theoretical Contributions and Practical Implications
This work contributes to strategy and AI research. First, we contribute to strategy research by showing how
value is created from AI-based BMs. We show the six value drivers of AI that create value and the case firms
draw their competitive advantage from the use of AI. Our model displays the complexity of AI-based value
creation and thus the underlying BMs. It shows that digital BMs increase the variety of value creation and
BM design options firms can build upon. Thereby we respond to calls for research on how AI impacts a
firm’s business logic (Weber et al. 2021). Second, we contribute to AI research by showing how AI creates
value and how it differs from previous digital innovations (Berente et al. 2021). The understanding of how
AI can contribute to value creation and competitive advantage justifies investments in AI projects. For
practice, we provide insights into the business value of AI. Understanding how AI creates value is essential
for firms to successfully transform their BMs. Based on our model of AI value creation, firms can detect
value creation potentials from using AI.
Limitations and Future Research
The chosen methodology leads to some limitations. First, AI applications are currently being developed at
a rapid pace, resulting in constant new opportunities for AI-based BMs (Wamba-Taguimdje et al. 2020;
Wodecki 2019). This, in turn, means that in the future it may be necessary to revisit the established model.
Second, due to the topic’s novelty, the data focus on start-up firms, even though already listed on stock
exchanges. Therefore, limited information was available on some of these firms. Third, the data were coded
using publicly available information, which meant that individual AI applications could not be considered
at the process level. It limits our analysis as we could not analyze the AI (data) models, algorithms, or other
internally used technology infrastructure that might affect the firm-internal value creation from AI. In our
future research, we will acknowledge these limitations, especially regarding the data collection, by collecting
primary data from the analyzed firms to validate our analysis.
Building on this work, several opportunities for future research arise. First, this work explored value
creation through AI at the BM level. However, not every firm uses AI at the BM level to improve operational
business processes (Wodecki 2019). Value creation at this operational level may be different from strategic
value creation at the BM level. Second, our cases show that AI is often linked to other digital innovations.
Quantum computing and faster mobile communications with 5G networks will soon enable even more AI
applications. Thus, value creation from AI can change as quickly as digital innovations emerge. Therefore,
future research should analyze new ways of value creation and, if necessary, to supplement our model, for
example with interdependencies between multiple digital innovations. Third, we identified an ecosystem as
an overarching value driver that influences value creation from AI. Future research could refine our model
to include the ecosystem perspective of how value can be co-created by multiple firms combining AI
technologies and capabilities. Fourth, we considered only the positive value created by AI, but its
implementation can also destroy value at the individual and organizational levels. Possible downsides
should be investigated.
Conclusion
The rapid development of AI applications offers firms entirely new opportunities to create value. This work
investigates the value creation potential of AI. To this end, a case-survey-based analysis of 61 firms and
their BMs yielded two key findings: first, value creation through AI can be mapped by six value drivers.
Second, we present a model of AI value creation displaying strong interdependencies between the individual
value drivers that need to be considered when designing AI-based BMs. This work thus contributes to a
more detailed understanding of how value can be created at the BM level through AI. Firms that are about
to transform to an AI-based BM can thus be supported in identifying the value drivers that are suitable for
them. Additionally, this work contributes to the AI research community's understanding of how AI can be
used to generate a competitive advantage.
Value Drivers of Artificial Intelligence
Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022
9
Acknowledgements
The authors would like to thank the track chairs, editors and all anonymous reviewers for their helpful
comments and suggestions. We thank the German Federal Ministry for Economic Affairs and Energy for
funding this research as part of the project 01MK20001B (Knowledge4Retail).
REFERENCES
Agrawal, A., Gans, J., and Goldfarb, A. 2019. The Economics of Artificial Intelligence: An Agenda. Chicago:
The University of Chicago Press.
Alsheibani, S.A., Cheung, Y., Messom, D.C., and Alhosni, M. 2020. "Winning Ai Strategy: Six-Steps to
Create Value from Artificial Intelligence," 26. Americas Conference on Information Systems (AMCIS),
Virtual, p. 11.
Amit, R., and Zott, C. 2001. "Value Creation in E-Business," Strategic Management Journal (22:6-7), pp.
493-520.
Amit, R., and Zott, C. 2012. "Creating Value through Business Model Innovation," MIT Sloan Management
Review (53), pp. 36-44.
Amshoff, B., Dülme, C., Echterfeld, J., and Gausemeier, J. 2015. "Business Model Patterns for Disruptive
Technologies," International Journal of Innovation Management (19:03), p. 1540002.
Benbya, H., Nan, N., Tanriverdi, H., Tanriverdi, H., and Yoo, Y. 2020. "Complexity and Information
Systems Research in the Emerging Digital World," MIS Quarterly (44:1), pp. 1-17.
Berente, N., Gu, B., Recker, J., and Santhanam, R. 2021. "Managing Artificial Intelligence," MIS Quarterly
(45:3).
Borges, A.F., Laurindo, F.J., Spínola, M.M., Gonçalves, R.F., and Mattos, C.A. 2021. "The Strategic Use of
Artificial Intelligence in the Digital Era: Systematic Literature Review and Future Research Directions,"
International Journal of Information Management (57), p. 102225.
Brynjolfsson, E., and Mcafee, A. 2017. "Artificial Intelligence, for Real," Harvard Business Review (1), pp.
1-31.
Chesbrough, H. 2007. "Business Model Innovation: It's Not Just About Technology Anymore," Strategy &
Leadership (35:6), pp. 12-17.
Chesbrough, H., Lettl, C., and Ritter, T. 2018. "Value Creation and Value Capture in Open Innovation: Value
Creation and Value Capture," Journal of Product Innovation Management (35:6), pp. 930-938.
Corbin, J., and Strauss, A. 1990. "Grounded Theory Research: Procedures, Canons, and Evaluative
Criteria," Qualitative Sociology (13:1), pp. 418–427.
Davenport, T.H., and Ronanki, R. 2018. "Artificial Intelligence for the Real World," Harvard Business
Review (96:1), pp. 108-116.
Gartner. 2019. "Gartner Says Ai Augmentation Will Create $2.9 Trillion of Business Value in 2021."
George, G., and Lin, Y. 2017. "Analytics, Innovation, and Organizational Adaptation," Innovation (19:1),
pp. 16-22.
Goodfellow, I., Bengio, Y., and Courville, A. 2016. Deep Learning. Cambridge, MA, USA: The MIT Press.
Grant, R.M. 1996. "Toward a Knowledge-Based Theory of the Firm," Strategic Management Journal (17:2),
pp. 109-122.
Güngör, H. 2020. "Creating Value with Artificial Intelligence: A Multi-Stakeholder Perspective," Journal of
Creating Value (6:1), pp. 72-85.
Hartmann, P., and Henkel, J. 2020. "The Rise of Corporate Science in Ai: Data as a Strategic Resource,"
Academy of Management Discoveries).
Hartmann, P.M. 2019. Profiting from Artificial Intelligence: Data as a Source of Competitive Advantage.
Norderstedt: BoD - Books on Demand.
Hein, A., Weking, J., Schreieck, M., Wiesche, M., Böhm, M., and Krcmar, H. 2019. "Value Co-Creation
Practices in Business-to-Business Platform Ecosystems," Electronic Markets (29:3), pp. 503-518.
Jöhnk, J., Weißert, M., and Wyrtki, K. 2021. "Ready or Not, Ai Comes—an Interview Study of Organizational
Ai Readiness Factors," Business & Information Systems Engineering (63:1), pp. 5-20.
Kaartemo, V., and Helkkula, A. 2018. "A Systematic Review of Artificial Intelligence and Robots in Value
Co-Creation: Current Status and Future Research Avenues," Journal of Creating Value (4:2), pp. 211-
228.
Value Drivers of Artificial Intelligence
Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022
10
Kitsios, F., and Kamariotou, M. 2021. "Artificial Intelligence and Business Strategy Towards Digital
Transformation: A Research Agenda," Sustainability (13:4).
Larsson, R. 1993. "Case Survey Methodology: Quantitative Analysis of Patterns across Case Studies,"
Academy of Management Journal (36:6), pp. 1515-1546.
Lee, G., DeLone, W.H., Tan, M., and Corrales, M. 2014. "Special Issue on Leveraging the Is Organization
for Business Value Creation," Journal of Information Technology (29:2), pp. 111-113.
Lee, J., Suh, T., Roy, D., and Baucus, M. 2019. "Emerging Technology and Business Model Innovation: The
Case of Artificial Intelligence," Journal of Open Innovation: Technology, Market, and Complexity
(5:3), p. 44.
Li, J., Li, M., Wang, X., and Thatcher, J.B. 2021. "Strategic Directions for Ai: The Role of Cios and Boards
of Directors," MIS Quarterly (45:3b), pp. 1603-1643.
Mikalef, P., and Gupta, M. 2021. "Artificial Intelligence Capability: Conceptualization, Measurement
Calibration, and Empirical Study on Its Impact on Organizational Creativity and Firm Performance,"
Information & Management (58:3), p. 103434.
Murphy, R. 2019. Introduction to Ai Robotics, (2 ed.). Cambridge, MA, USA: MIT Press.
Nowacki, C. 2019. "What Is the User Value of Ai? A Taxonomy Based on Ai Startups in France in 2019,"
SSRN Electronic Journal).
Osterwalder, A., and Pigneur, Y. 2010. Business Model Generation: A Handbook for Visionaries, Game
Changers, and Challengers. Amsterdam: Wiley.
Raisch, S., and Krakowski, S. 2021. "Artificial Intelligence and Management: The Automation–
Augmentation Paradox," Academy of Management Review (46:1), pp. 192-210.
Ransbotham, S., Gerbert, P., and Kiron, D. 2017. "Reshaping Business with Artificial Intelligence," MIT
Sloan Management Review.
Reis, C., Ruivo, P., Oliveira, T., and Faroleiro, P. 2020. "Assessing the Drivers of Machine Learning Business
Value," Journal of Business Research (117), pp. 232-243.
Schmidt, R., Möhring, M., and Zimmermann, A. 2020. "Value Creation in Connectionist Artificial
Intelligence – a Research Agenda," 26. Americas Conference on Information Systems (AMCIS),
Virtual, p. 11.
Schumpeter, J. 1934. The Theory of Economic Development: An Inquiry into Profits, Capital, Credit,
Interest, and the Business Cycle. London, UK: Oxford University Press.
Someh, I., Wixom, B., and Zutavern, A. 2020. "Overcoming Organizational Obstacles to Artificial
Intelligence Project Adoption: Propositions for Research," 53. Hawaii International Conference on
System Sciences (HICSS), Honolulu, HI, USA.
Steininger, D.M. 2019. "Linking Information Systems and Entrepreneurship: A Review and Agenda for It-
Associated and Digital Entrepreneurship Research," Information Systems Journal (29:2), pp. 363-407.
Teece, D.J. 2010. "Business Models, Business Strategy and Innovation," Long Range Planning (43:2-3),
pp. 172-194.
Teece, D.J. 2018. "Profiting from Innovation in the Digital Economy: Enabling Technologies, Standards,
and Licensing Models in the Wireless World," Research Policy (47:8), pp. 1367-1387.
Veit, D., Clemons, E., Benlian, A., Buxmann, P., Hess, T., Kundisch, D., Leimeister, J.M., Loos, P., and
Spann, M. 2014. "Business Models," Business & Information Systems Engineering (6:1), pp. 45-53.
Vial, G. 2019. "Understanding Digital Transformation: A Review and a Research Agenda," The Journal of
Strategic Information Systems (28:2), pp. 118-144.
Wamba-Taguimdje, S.-L., Fosso Wamba, S., Kala Kamdjoug, J.R., and Tchatchouang Wanko, C.E. 2020.
"Influence of Artificial Intelligence (Ai) on Firm Performance: The Business Value of Ai-Based
Transformation Projects," Business Process Management Journal (26:7), pp. 1893-1924.
Wang, N., Liang, H., Zhong, W., Xue, Y., and Xiao, J. 2014. "Resource Structuring or Capability Building?
An Empirical Study of the Business Value of Information Technology," Journal of Management
Information Systems (29:2), pp. 325-367.
Weber, M., Beutter, M., Weking, J., Böhm, M., and Krcmar, H. 2021. "Ai Startup Business Models,"
Business & Information Systems Engineering).
West, D.M. 2018. The Future of Work: Robots, Ai, and Automation. Washington, D.C: Brookings
Institution Press.
Wodecki, A. 2019. Artificial Intelligence in Value Creation: Improving Competitive Advantage. Cham:
Springer International Publishing.
Yin, R.K. 2017. Case Study Research and Applications: Design and Methods, (6th edition ed.). Los Angeles,
USA: SAGE Publishing.
