Content uploaded by Umar Mir
Author content
All content in this area was uploaded by Umar Mir on May 15, 2022
Content may be subject to copyright.
AI-enabled digital identity –inputs
for stakeholders and policymakers
Umar Mir,Arpan Kumar Kumar Kar and Manmohan Prasad Gupta
Indian Institute of Technology Delhi, New Delhi, India
Abstract
Purpose –This conceptual article’s primary aim is to identify the significant stakeholders of the digital
identity system (DIS) and then highlight the impact of artificial intelligence (AI) on each of the identified
stakeholders. It also recommends vital points that could be considered by policymakers while developing
technology-related policies for effective DIS.
Design/methodology/approach –This article uses stakeholder methodology and design theory (DT) as
a primary theoretical lens along with the innovation diffusion theory (IDT) as a sub-theory. This article is
based on the analysis of existing literature that mainly comprises academic literature, official reports, white
papers and publicly available domain experts’interviews.
Findings –The study identified six significant stakeholders, i.e. government, citizens, infrastructure
providers, identity providers (IdP), judiciary and relying parties (RPs) of the DIS from the secondary data.
Also, the role of IdP becomes insignificant in the context of AI-enabled digital identity systems (AIeDIS). The
findings depict that AIeDIS can positively impact the DIS stakeholders by solving a range of problems such
as identity theft, unauthorised access and credential misuse, and will also open a possibility of new ways to
empower all the stakeholders.
Research limitations/implications –The study is based on secondary data and has considered DIS
stakeholders from a generic perspective. Incorporating expert opinion and empirical validation of the
hypothesis could derive more specific and context-aware insights.
Practical implications –The study could facilitate stakeholders to enrich further their understanding
and significance of developing sustainable and future-ready DIS by highlighting the impact of AI on the
digital identity ecosystem.
Originality/value –To the best of the authors’knowledge, this article is the first of its kind that has used
stakeholder theory, DT and IDT to explain the design and developmental phenomenon of AIeDIS. A list of six
significant stakeholders of DIS, i.e. government, citizens, infrastructure providers, IdP, judiciary and RP, is
identified through comprehensive literature analysis.
Keywords Stakeholder theory, Biometrics, Artificial intelligence, Design theory,
Innovation diffusion theory, Digital identity
Paper type Conceptual paper
1. Introduction
Identity (ID), in general, is a very complex concept that has multiple dimensions to it. The
research community have explored the concept of “identity”from various disciplines such
as psychology, information technology, philosophy, sociology, anthropology, politics and
psychoanalysis (Camp, 2004;Ellemers et al., 2002;Jung and Mittal, 2020). The study of
“identity”at the core is about the answers to basic questions such as “who are you?”or “who
am I?”It is composed of elements that distinguish an individual uniquely from the rest of the
population (Olson, 2015). Digital identity (DI), sometimes also referred to as electronic
Funding: This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.
AI-enabled
digital identity
Received 12 September2020
Revised 15 December2020
17 March 2021
29 May 2021
Accepted 31 May 2021
Journal of Science and Technology
Policy Management
© Emerald Publishing Limited
2053-4620
DOI 10.1108/JSTPM-09-2020-0134
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/2053-4620.htm
identity (eID), has revolutionised the public service delivery mechanism, e.g. public
distribution system (PDS) in India. DI provides a niche opportunity, especially to developing
countries, in accelerating the overall developmental programs(Mir et al., 2020b). DI has been
defined as the “collection of individual attributes that describe an entity and determine the
transactions in which that entity can participate”(Mcwaters, 2016). In another study, DI is
defined as a “set of claims made by one digital subject (e.g. a user) about itself or about
another digital subject”(Cameron, 2005). In simple terms, DI is the online repository of
information about a particular individual that is accumulated from multiple data sources. DI, in
general, impacts various socio-economic aspects of an individual, as shown in Figure 1.
Rapidgrowthinthefield of Information and Communication Technology (ICT) has
widened the scope for the development of people (Albiman and Sulong, 2016;Chatterjee and
Kar, 2018). In the recent past, ICT researchers have explored the impact of disruptive
technologies such as artificial intelligence (AI) on the growth of nations in general and on a
citizen in particular (Agrawal et al., 2018;Helbing et al., 2019;Wirtz et al., 2019). Expectations
from emerging technologies are very high considering its potential benefits in the public and
private sectors and society in general in terms of better service quality, reducing lead time and
better decision-making capabilities (Toll et al., 2020). Disruptive technologies such as
blockchain, cloud computing and AI have drastically transformed the nature of transactions
from offline space to online space, and hence opening a new dimension wherein completing a
transaction invisibly is almost impossible (Rifkin, 2001). In the past decade, AI has evolved
rapidly, which has resulted in its application across sectors (Mir et al., 2020c). The top four
sectors, i.e. retail, logistics, travel and health care, are leading in AI application with an impact of
around $400–$600bn approximately in each sector (McKinsey, 2017a). Apart from assisting
humans in medical treatment (Amato et al., 2013), manufacturing self-driving vehicles and safety
systems (Tajitsu, 2016), developing context-aware virtual assistants (Waters, 2015)and
enhancing entertainment industry (Bates, 1992), AI is also being used for identifying fraud and
help take proactive measure to mitigate possible attacks (Ryman-Tubb et al.,2018).
Providing identity that could be used for an individual’s well-being is a highly complex
process (Mir et al., 2020a). This has been noticed by various bodies that operate at global
levels such as the United Nation (UN) and the World Bank. In its recently released
sustainable development goals (SDGs), UN has set a milestone of providing a legal identity
to all individuals by 2030 (SDG-16). Also, the World Bank has initiated the Identity for
Development (ID4D) project that aims to provide identity to individuals and deliver DI-
enabled services to every individual. Even after many conscious efforts made by multiple
Figure 1.
Impact of DI on socio-
economic aspects
Governance Financial Inclusion
Gender Equality
Access to Healthcare
Access to Educaon
Access to PDS
Digital
Identy
Source: Goode (2010); Kar 2020; McKinsey (2019);
Mcwaters (2016)
JSTPM
organisations, only 3% of the countries possess an identity system that could be used in
both online and offline service delivery. Majority of the identity solutions at present are
functional, which means they are developed for a specific use case such as driving licence,
PAN card and voter ID (Mir et al., 2020a). Governments have already used AI systems as a
solution to bring efficiency, and enable efficient utilisation of resources in the public sector
(Toll et al., 2020). AI could be used in public policy at various stages such as agenda-setting,
policy formulation, decision-making, policy implementation and evaluation. Each stage has
its own set of challenges that needs to be addressed beforehand such that full potential of the
technology could be leveraged (Dunn, 2017).
DI is the cornerstone of the digital economy and is believed to significantly impact digital
development (Al-Khouri, 2014;BankWorld, 2016;OCED, 2011). This has led various
governments to opt for ICT-based solutions to solve the “invisible population”problem (i.e.
people who do not possess any form of legal identity), improve public service delivery and
leverage the benefits of the digital economy. Demands for such identity systems are increasing
when there is a technological revolution happening on the supply side (Atick, 2016). Disruptive
technologies such as AI, IoT and big data have significantly impacted almost all sectors. These
innovations could be game changers if used properly. Technologies that are used in digital
identity systems (DISs) are evolving at a rapid pace. This requires some serious efforts to
anticipate the evolution of DIS infrastructure in the next few decades. Any miss-out in this
direction could result in an ineffective DIS that will incur substantial socio-economic
losses (Beduschi et al., 2017). There is a dire need for focused research towards
developing the appropriate DI system that is future-proof, secure and trustworthy and
could be used for identificationandservicedeliverybybothpublicandprivateentities
(Bhandari et al., 2020). This study attempts to highlight the significance of a disruptive
technology such as AI in the identification phenomena. Specifically, in this conceptual
article, which is based on the secondary data, we try to analyse how AI could be
leveraged for identification and try to answer the following research questions:
RQ1. Who will be the stakeholders of AI-enabled digital identity systems?
RQ2. How can AI enhance capabilities of these stakeholders?
The subsequent sections of this article are organised as follows. Section 2 covers the
literature review. Section 3 focuses on the methodology followed in this research, and
theoretical linkages are explained in Section 4. Findings are presented in Section 5, followed
by discussion, contribution and limitations and future research directions in Section 6.
Finally, Section 7 covers the conclusion part.
2. Literature review
AI is not unusual to the government. While its utility in domains such as defence,
surveillance and intelligence have been accepted, it also has been used to lessen heavy
tasks. The recent advancements in technology, especially networking, data processing and
data storage capabilities, have opened gates for many opportunities and have broadened
the utility dimension of the technology (Dwivedi et al., 2019;Kar, 2016). This has led
various countries to be part of the digital revolution by taking the first step in this direction
and formulating a national AI strategy, as shown in Table 1. The readiness index
highlights how prepared government is for the inclusion of AI in governance processes
(Miller and Stirling, 2019).
AI-enabled
digital identity
AI is considered a specialised science category that facilitates machines to identify a suitable
solution for solving complex problems as normal human beings do. Various researchers
from different lenses have defined AI. For example, Rai (2018) defined AI as:
[...] the ability of a machine to perform cognitive functions that we associate with human minds,
such as perceiving, reasoning, learning, interacting with the environment, problem solving,
decision-making, and even demonstrating creativity.
Some defined it as a technology that behaves intelligently, imitates humans or possesses
cognitive capabilities like humans (Flasi
nski, 2016;Robles Carrillo, 2020;Warwick, 2012).
Irrespective of which definition one uses, there is some level of agreeability among the AI
research community that AI is a mix of multiple domains that include computer science,
philosophy, linguistics, mathematics and psychology.
AI has found application in various domains such as public sector (Pan, 2016;Simon,
2019), health care (Winter and Davidson, 2019), transportation (W.Sadek, 2007), hospitality
(Buhalis et al.,2019;Li et al., 2019), education (Passonneau et al., 2017), justice (Rissland et al.,
2003), entertainment (Bowman and Banks, 2019) and retailing (Weber and Schütte, 2019).
Some researchers have focused on the specific utility of AI in a specific area (Mir et al.,
2020c). One of the most significant contributions of AI in government is developing
intelligent agents, also called Bots. Bots have drastically improved the communication
Table 1.
National AI strategy
landscape
Year Country
Govt.
readiness
rank AI initiative
Mar-2017 Canada 6 Pan-Canadian artificial intelligence strategy
Mar-2017 Japan 10 Artificial intelligence technology strategy and also included
AI in its “integrated innovation strategy”
May-2017 Singapore 1 National AI strategy and AI Singapore
July-2017 China 20 A next-generation artificial intelligence development plan
Oct-2017 UAE 19 National strategy for AI
Dec-2017 Finland 5 Finland’s age of artificial intelligence
Jan-2018 Kenya 52 Blockchain and artificial intelligence task force
Jan-2018 Taiwan 41 Taiwan AI action plan
Jan-2018 Denmark 9 Strategy for Denmark’s digital growth
Jan-2018 Estonia 23 Estonia’s AI task force
Mar-2018 Italy 15 AI task force
Mar-2018 France 8 National strategy for AI called “AI for humanity”, which is
outlined in the “Villani report”
Apr-2018 Tunisia 54 AI task force and steering committee
Apr-2018 UK 2 Sector deal for AI to advance the UK’s ambitions in AI
May-2018 Australia 11 AI policy
May-2018 USA 4 American AI initiative
May-2018 South Korea 26 Artificial intelligence information industry Development strategy
May-2018 Sweden 6 National approach for artificial intelligence
June-2018 India 17 National strategy for artificial intelligence #AIforAll
June-2018 Mexico 32 Towards an AI strategy in Mexico: harnessing the AI revolution
Aug-2018 Saudi Arabia 78 Saudi data and artificial intelligence authority
Nov-2018 Germany 3 Artificial intelligence strategy
Apr-2019 Lithuania 37 Lithuanian artificial intelligence strategy
Oct-2019 The Netherlands 14 Strategic action plan for artificial intelligence
Oct-2019 Russia 29 National AI strategy
Jan-2020 Norway 12 National strategy for artificial intelligence
JSTPM
phenomena between citizens and governments (Androutsopoulou et al., 2019). Studies such
as Pantano and Pizzi (2020) analysed AI’s application, primarily covering the development
of specialised chatbots for assisting customers in online retailing. From the governance
perspective, AI is believed to have a multi-fold impact on people, governments and public
service delivery. Researchers have explored the utility of AIin different contexts, e.g. Ayoub
and Payne (2016) discuss the impact of AI-assisted strategic decision-making, Brynjolfsson
and Mitchell (2017) highlight the implication of AI on the workforce, considering both
pros and cons of the technology, Sharma et al. (2020) studied the application of AI in public
sector and Ku and Leroy (2014) highlight the role AI in crime detection. Apart from all the
promising benefits of AI, it also brings some peculiar challenges. For example, Sun and
Medaglia (2019) analyse the challenges faced by various stakeholders in the adoption of AI
in health care.
The increasing demands primarily drive the digitalisation of public services and other
types of transactions, including financial in terms of cost reduction, efficient service delivery
and reduction in fraudulent transactions. This needs an intelligent identity system that not
only identifies an individual uniquely but also addresses the requirement as mentioned
above for a better governance experience –AI-enabled digital identity systems (AIeDIS). It
is proven in various studies that AI-enabled systems are good at detecting frauds (Bolton
and Hand, 2002), improve user experience through Bots (Duijst, 2017) and identify an
individual based on their biological traits such as biometrics (Liang et al., 2020). This leads
us to believe that AIeDIS can address the needs of both an individual and government and
improve the overall experience of both entities. Stakeholders of AIeDIS, which is a mix of
technology and non-technology entities, are the amalgam of stakeholders of AI and
e-government systems. Some of the primary stakeholders of AI and e-government are
citizen, government, technology providers, NGOs, judiciary and service providers (Al-
Khouri, 2014;Ashaye and Irani, 2019;Mir et al., 2020b,2020c;Rowley, 2011).
AI plays an essential role in the identification systems and has better accuracy over
traditional identification processes. With the help of AI, facial recognition can detect
between actual human and spoofed ones with higher accuracy during the KYC process.
With the continuous improvements in the AI algorithms, AI-enabled systems can detect a
forgery in identity documents with better precision than the manual screening process. Such
systems not only bring higher accuracy in the verification process but also improve identity
verification by multi-fold. There are multiple ways by which AI can improve DIS in terms of
speed, accuracy, efficiency and utility. AI plays a pivotal role in determining real identity,
detects fake documents, faster identity verification and is very hard to circumvent.
Researchers (Phiri et al., 2011a) proposed an AI-enabled multifactor authentication approach
(MFA) for a secure authentication system, wherein biometric, pseudo-metric and device
metric factors are used for authentication. Phiri et al. (2011b) used the information fusion
technique for MFA. They mined multiple data sources such as financial institutions,
education systems, health-care systems, social networks and questionnaires to compile
identity attributes. In another study (Cole, 1991), the author introduces the concept of a
virtual person based on virtual machines and concludes that no digital machine could
understand language and how artificial minds are possible. From the identification
perspective, we did not find any significant work that has studied the impact of AI on the
DIS ecosystem. Some of the notable studies that have used AI for different identification
forms are shown in Table 2.
In general, AI for authentication services is growing, but it is still in its inception phase. It
is expected that researchers will explore multiple dimensions of AI specific to identification
services in the near future, considering the development and demand in AI and
AI-enabled
digital identity
identification, respectively. In this study, we try to address this gap by focusing on the
application of AI for the digital identification and its implications on the DIS stakeholders.
3. Methodology
The study used the stakeholder methodology to identify the primary stakeholders that
AIeDIS could impact. ST is different from others. It is very effective in addressing the needs
and interests of key entities during the planning process, which impacts the system’s overall
performance. It is a procedure to systematically gather and analyse information to identify
actors or stakeholders that significantly impact the overall system. ST is very effective in
this study because it enables policymakers and practitioners to identify the key entities and
assess their position and significance within the system and develop relevant policies and
implementation plans accordingly. Having a prior understanding of key entities and their
impact on the overall system increases the program’s chances of being successful.
The systems approach is another alternative that could have been used in this study. In
the systems approach, the focus of analysis is mainly on the holistic aspects of the system. It
analyses the interaction and inter-relationship mechanism among the parts of the system
and their impact on each other to understand the functioning of the overall system. In this
study, ST is chosen over systems approach because the intent is to identify the core entities
of the system that could be impacted by AI’s inclusion at an individual level rather than
analysing the entities’intra-dependence relation. Another important reason is that the
outcome of ST enables us to focus on each stakeholder explicitly from the design
perspective, thereby satisfying design theory (DT)’s initial requirements.
All these stakeholders were identified from an in-depth analysis of various secondary
data sources that included research articles, white papers from various government and
non-government entities, news articles and publicly available interviews of experts. To
explore the relevant literature for this study, we used a rigorous methodology opted by
Abedin (2013) for identifying resources. We queried Scopus for relevant literature, which is
one of the largest abstracts repositories. Although most of the literature was taken from
Scopus, we also used Google Scholar as our second-line data source. Articles were identified
by using three key search keywords, i.e. “Artificial Intelligence”,“Digital Identity”and
“Electronic Identity”. In the initial stage, 142 articles were identified after dropping
duplicates and entries with missing original texts. In the next stage, we read the abstract
and conclusion section of each paper and filtered out 86 articles that were deemed irrelevant
Table 2.
Application of AI for
identification
Author(s) Application of AI
Strich et al. (2021) Impact of substitutive decision-making AI systems on employees’professional
identity
Phiri et al. (2011a) Proposed AI-based multifactor authentication mechanism
Cole (1991) Highlighted possibility of artificial minds using emerging technologies
Verma et al. (2019) Proposed a mechanism to predicted national identity of students
Lalmuanawma et al.
(2020)
Scope of AI for Covid-19 patient tracking
Horowitz et al. (2018) Identification of outliers in national security context
Agbinya (2019) Application of AI in identity management systems
Zanzotto (2019) Application of human-assisted AI systems for identifying people
Power (2016) Identification of criminals in public places
de Vries (2010) Impact on one’s identity by machine-enabled profiling
Phiri et al. (2011b) Mined multiple data sources using AI for compiling identity
JSTPM
for this study. Further, 37 more articles were dropped that were identified as loosely relevant
for this study after reading the complete paper. Hence, in total, 123 articles were dropped
based on the exclusion criteria such as not relevant in the context of identification, missing
full-text and published in low-quality journals. Finally, a total of 31 articles, which include
19 academic papers and 12 additional documents in the form of reports and white papers,
were considered for further analysis. The steps followed during literature selection are also
shown in Figure 2. Each article was thoroughly analysed by the authors separately. During
consensus building, each author’sfindings were collated, and a total of ten stakeholders
were identified in the initial phase from the literature analysis. In the subsequent phase,
some of the closely related stakeholders were clubbed together and were replaced by a single
stakeholder category, e.g. platform and technology is replaced by infrastructure; biometric
data collectors and on-boarding of identity data are replaced by identity providers (IdPs);
and non-government organisations (NGOs), regulatory body and government were clubbed
into government. Finally, six significant stakeholders were identified (see Table 3) and are
used for further analysis.
ST is a well-studied and extensively applied methodology in different domains of
research that focus on the category which are essential for the organisation (Mishra and
Mishra, 2013;Singh et al.,2017). Its importance, especially in the e-governance domain, has
Figure 2.
Steps followed in
identifying relevant
literature
Table 3.
Stakeholders of DIS
Stakeholder Reference
Government Dixon (2017),McKinsey (2019);Mcwaters (2016);Mir et al. (2020b);Moavenzadeh and de
Maar (2018),Open Identity Exchange (2018)
Citizens Clark et al. (2016),Mcwaters (2016);Mir et al. (2020b);Moavenzadeh and de Maar (2018),
Ronald et al. (2017);Schoemaker et al. (2020)
Infrastructure
providers
Goodell and Aste (2019),ITU-T FG-DFS (2017);Laurent and Bouzefrane (2015);Mir et al.
(2020b);Schoemaker et al. (2020)
ID providers Clark et al. (2016),Knight and Saxby (2014);Mcwaters (2016);Mir et al. (2020b);Sharma
(2016)
Judiciary Bhandari et al. (2020),Dixon (2017);Open Identity Exchange (2018)
Relying parties Bhandari et al. (2020),Clark et al. (2016);Goodell and Aste (2019),Mcwaters (2016)
AI-enabled
digital identity
been extensively studied in the literature (Janssen and Estevez, 2013;Van De Kaa et al.,
2018). Studies have observed that stakeholders are key factors in the success or failure of
implementing new technologies (Rahman and Ko, 2013). Ema et al. (2016) have used a
stakeholder approach to study the opinions of stakeholder concerning AI and human–
machine interactions. ST is relevant in the context of this study as the aim of this study is to
analyse the impact of AI technology on DI systems andwhich entities are getting affected in
the existing identity system platform.
4. Theoretical lens
This study uses DT as the primary theoretical lens. DT adopted in this study is based on IS
design theory (Walls et al., 1992) and DI design theory (Mir et al., 2020b). DT requires initial
constructs to be derived using kernel theory. Hence, in this study, we have used two well-
established theories, i.e. ST and innovation diffusion theory (IDT), as fundamental kernel
theories. Each theory holds significant importance in the context of this study. The purpose
of using ST is to identify the stakeholders of AIeDIS and IDT to identify the critical
dimensions that play a vital role in adopting innovative technologies. The application and
organisation of theories used are explained below.
4.1 Design theory
Literature is indicative of the acceptance and implementation of DT in various scenarios
(Hatchuel et al.,2016). Different thoughts of school have defined DT differently, e.g. Walls
(1992) highlights the prescriptive nature of theory, Goldkuhl (2004) focuses on practical
aspects of it and Gregor and Jones (2007) stress on the DT as a basis for action. From the IS
perspective, Walls’IS DT has been widely accepted and followed by the IS community.
Walls divided DT for information systems into two major components, “Design Product”
and “Design Process”and defined DT for information systems as “a prescriptive theory
based on theoretical underpinnings which says how a design process can be carried out in a
way which is both effective and feasible”(Mir et al., 2020b).
DT holds two major advantages that make it more suitable for this study: 1) Its ability to
highlight the significance of stakeholders in the system; 2) Its ability to verify both product
and processes followed. This article applies DT from the prescriptive lens and explains the
design aspects of AIeDIS (see Figure 3). Next, we brieflydefine six steps of DT followed in
this study.
Identification of stakeholders: ST, originally proposed by Freeman (2015), is widely
accepted and actively applied in different research domains because of its simplicity and
effectiveness in analysing the significance of entities in a system. It focuses on the
relationships among different groups of actors both within and outside of the system. ST
helps in identifying who are the key actors within the system. ST in information system
discipline is defined as:
[...] all those who have a practical concern for the effective application of new technologies, and
who are in a position to take or to influence decisions about why and how they are used (Mishra
and Mishra, 2013).
Some of the main advantages of ST are ensuring transparency, accountability, credibility
and enhancing quality (Motu’apuaka et al., 2015). In this study, ST is used to identify
significant stakeholders in AIeDIS. Data was collected mostly from DI literature mainly
related to India’sDI–Aadhaar. Aadhaar is world’s largest biometric DI scheme based on
the number of enrolments (Mir et al., 2020b) and AI frameworks (Mir et al., 2020c). The focus
was to include all the significant stakeholders of the DIS.
JSTPM
Meta-requirements: This component identifies the list of features an AIeDIS should have,
which acts as essential requirements for applying theory. Meta-requirements (MRs) are
listed by analysing the existing DIS literature gaps and rising concerns among the
developing and less developing countries. Initially, an extensive list of requirements was
identified, which were later classified corresponding to six significant stakeholders of the
AIeDIS, as shown in Table 3.
Meta-design: Meta-design (MD) constitutes the set of recommendations from policy and
technology perspectives corresponding to each MR. Recommendations are made keeping in
view the power and applicability of AI in DIS cycle. The design recommendations focus on
fool-proof identification, convenience, data protection and robustness of the overall system.
MD corresponding to each MR is listed in Table 5.
Innovation diffusion theory: The study uses IDT as a kernel theory, which acts as guiding
principle for DM. IDT is a seminal work in communications discipline originated from
Psychology domain. IDT is used to explain how innovations are adopted among the target
audience Although Gabriel Trade initially explored it, the final theory came into existence in
1962 by Everett Rogers (Bhattacherjee, 2020). Application of IDT in IS and its sub-domains
have grown immensely in the past two decades (Lim et al.,2013). IDT in the context of this
study is used to enhance the probability of adoption of AIeDIS among stakeholders. Five
major innovation characteristics taken from IDT are compatibility, complexity, relative
advantage, observability and trialability. Compatibility is the extent to which innovation (AI
in the context of this study) is considered as consistent with the existing systems, user
experience and needs of the stakeholders. Complexity is the degree of the perceived
difficulty in terms of developing AIeDIS, and ease of use for stakeholders. Relative
advantage is the realisation of potential benefits that a new system based on innovation
brings over the existing system. This is considered to have a significant impact on the
adoption level of an innovation-based idea. Observability is the extent to which the target
population observes the outcome of innovation. Trialability implies the possibility of testing
an innovation before the final system. It is primarily achieved by developing a prototype of
the intended system.
Figure 3.
Components of
AIeDIS design theory
using stakeholder
and innovation
diffusion theory
adapted from Mir
et al. (2020b) and
Walls et al. (1992)
AI-enabled
digital identity
Design method: Design method provides possible scenarios highlighting how MD could
be achieved. In this conceptual article, we used DT from the prescriptive lens. We explained
how AIeDIS MR could be met by incorporating the proposed MD recommendations. This
section is the process guidelines that could be followed to achieve construction of MD
components of AIeDIS.
Meta-validation: Meta-validation is concerned with the testing of the overall system and
has two primary checks to make 1) if MD satisfies MR and 2) if the design method results in
the MD compatible artefacts. The MR and MD need to be tested empirically only after
developing a functional prototype. Guidelines for testing AIeDISare highlighted inTable 4.
5. Findings
This section highlights the impact of AI on the stakeholders of an identity system. Table 5
depicts the relationship between the AIeDIS requirements and the corresponding design
demands, which is further explained in this section. Shifting of functionalities from one
stakeholder to another, adding/removing functionalities and justification regarding
removing some existing stakeholder are considered an impact in the subsequent sections of
this article. Table 6 presents a snapshot of the degree of impact of AIeDIS on the
stakeholders corresponding to five IDT characteristics that have been identified as critical
for adoption (Bhattacherjee, 2020).
5.1 Government
AI’s application first started when the Enygma machine was used for decoding encrypted
Nazi’s communications during World War II. Since then, digital technologies such as AI
have evolved significantly in terms of effectiveness and efficiency. At present, AI has
touched almost all the major sectors such as automobile, fintech, education, health care and
logistics. Implementation of AI in delivering eGovernment services is picking up rapidly.
Researchers have used AI as a baseline technology to develop domain-specific solutions.
Androutsopoulou et al. (2019) have developed a chatbot using AI to communicate between
citizens and government. This chatbot not only enables citizens to interact with the
government but also facilitates information seeking in a native language and conducting an
online transaction.
Policymaking is a daunting task for any government. It has an impact on the social,
economic and environmental aspects of society. AI could be implemented in the public
service sector for developing effective policies (Lauterbach, 2019). As per the recent report
from World Economic Forum, systematic adoption of AI in government faces five major
hindrances related to the use of data, AI skills, AI ecosystem and legacy culture and
procurement mechanism (Santeli and Gerdon, 2019). There are different types of AI-
associated techniques that could help in solving some of these issues. For example, from a
Table 4.
Testing of AIeDIS
1 It feasible to develop AIeDIS
2 It is feasible to address gaps of existing DIS via AIeDIS
It is feasible to develop completely automatic identification system with present available technologies
3 It is feasible to develop consent-based DIS
4 AIeDIS should be compatible with existing DIS
5 It is possible to develop AIeDIS with minimum amount of personal data
6 It is feasible to develop laws exclusively for AI systems
7 AIeDIS prototype helps achieve MR
JSTPM
policy perspective, data mining, decision support systems, game theory and simulations
techniques could be used in policy development (Milano et al.,2014).
Identifying people has been a complicated task for the governments to tackle various
pressing issues such as illegal migrations, terrorism, citizenship and fraudsters’detection.
These issues are not new and have been there for ages. Governments in the past have taken
measures to deal with such issues by identifying people through various means such as
Table 5.
Design theory
constructs
Stakeholder Meta-requirements Meta-design Explanation
Government MR-1: AIeDIS should
support real-time
identification
MD-1: Automatic
identification at entry
and exit points
Use of biometric scanning
technologies at entry and exit points
MR-2: AIeDIS should
support citizen-centric
policy development
MD-2: Consent-based
analysis of non-personal
data
Opt-in approach prior to any
data collection and processing
activity
MR-3: AIeDIS should be
affordable for masses
MD-3: Use of commodity
technologies and
availability of
government subsidies
Apply cost reduction techniques
at both product and process
level; support for plug-n-play
development approach with
focus on reusability
Citizens MR-4: AIeDIS should
enhance better access to
public and private
services
MD-4: Uniform
identification platform
for access control
mechanism
24 7 easy-to-access platform
with yes–no authentication; focus
on on-boarding of service
providers
MR-5: AIeDIS should
mitigate identity theft
MD-5: zero cognitive load Issues no physical cards, unique
numbers, username etc.; use
biometrics of a person only
MR-6 AIeDIS should
support both monetary
and non-monetary
transactions
MD-6: Common
minimum identification
standards
Exclusively yes–no
authentication will shun the need
of transaction constrained
authentication
Infrastructure
providers
MR-7: AIeDIS should
support plug-n-play of
AI-friendly technologies
MD-7: Quality control
board
Standardised techniques and
technologies will enhance
seamless integration
ID providers MR-8: AIeDIS should
minimise human
intervention in
identification process
MD-8: Induction of
intelligent systems
Use of sensors for data collection
removes the need of human
intervention
Judiciary MR-9: Need for AI-
compatible laws and
regulations
MD-9: Inclusion of
experts in law-making
bodies
Domain experts consider
technology-specific issues that
could have significant impact in
developing technology-related
policies
MR-10: AIeDIS should
support in finalising
verdict
MD-10: AI-compatible
data repositories
Better decision-making could be
facilitated by analysing all the
relevant data that must be in
machine readable form
Relying
parties
MR-11: AIeDIS should
support quick and hassle-
free identification
MD-11: Automatic
identification
Fully technology-assisted
authentication could be free from
humanly bottlenecks
MR-12: AIeDIS should
support customer-centric
services to legitimate
users only
MD-12: Consent-based
analysis of non-personal
data for service
recommendation
Exclusive consent for data
processing could facilitate better
access control and user-centric
services recommendations
AI-enabled
digital identity
domicile certificates, national IDs, unique identity credentials and undertaking by
competent authorities. However, all these measures have one or the other loopholes that
have been exploited to gain unauthorised benefits by the people with ill intentions (Newman
and McNally, 2005). Earlier identity proofs were formulated primarily for verifying people in
physical space. As digital technologies evolved and disrupted almost all sections of society,
there is a need for an identity suitable for both online and offline space as earlier identity
proofs posed serious challenges (Allison et al., 2005;Belk, 2013).
AI could play a significant role in identifying people in the online space. The
advancements in AI subfields (machine learning, image processing, video processing and
pattern recognition) could solve a diverse set of need-based problems. For example, real-time
image and video processing could be used to identify suspected criminals, rash driving and
unauthorised trespassing by saving time and efforts. Face recognition will enable
governments to fool-proof attendance systems at entry and exit points. It will also help in
case of theft and illegal access. AIeDIS could achieve all these and much more. Some
countries such as China have already started using AI for various use cases such as home
security, border security and criminal identification; however, an identity system that is
fully AI-based is still missing in the present times (Larson, 2018). Such an identity system
will lead towards smart governance that will be less costly, removes human errors and
checks on corruption (de Sousa et al.,2019). Besides, it will also facilitate personalised
service delivery with full support for interoperability acrosssectors.
5.2 Citizens
Individuals are using AI services in their routine activities directly or indirectly, e.g. online
shopping, reading news, using social media, etc. AI-based technologies have been applied in
different use cases solving a diverse set of problems concerning the general public. Liu et al.
(2010) have highlighted how AI could be leveraged in mobile learning and identified five
specific application of AI used in mobile learning in terms of expert systems, decision
support systems, information retrieval system, induct-learning system and intelligent
hardware. Studies such as Shank et al. (2019) analysed the people’s perception towards AI
systems wherein users perceived possession of mental capabilities in AI systems during the
interaction. Irrespective of different types of interactions with AI systems, people develop
emotional connect when perceiving mind in AI. Authors classify these emotions into seven
categories, i.e. surprise, amazement, amusement, unease, happiness, disappointment and
confusion, and analyse their impact in developing these mental perceptions.
Perks of AI-enabled solutions must be available to all. They should be easy to use so that
the majority of the population could leverage them. It has the potential to break some of the
long lingering problems of the society such as inequalities in education, health care,
financial services and human rights. Identity has been one of the primary reasons for such
problems. According to the World Bank report, Desai et al. (2018) at present, there are more
than 1 billion people globally without any legal identity. In such circumstances, using AI for
Table 6.
Degree of impact of
AIeDIS on
stakeholders
Government Citizens Infrastructure Judiciary Relying parties
Compatibility Low Low Medium Low Low
Complexity High Low High Low Medium
Relative advantage High High Moderate Moderate High
Observability High High High Medium Medium
Trialability High Low Low Medium Low
JSTPM
providing identity to people will solve many problems, especially in developing countries
where the marginalised population is on the higher side and at high risk. Identity based on
the amalgamation of AI and biometrics is not very easy to compromise. It enables an
individual to perform transactions and avail various services confidently without worrying
much about frauds. Further, such an identity system not only saves time and is hassle-free but
also saves costs that otherwise would have been incurred by an individual in case of an offline
identity system. The best example of such an identity system is the immigration checking
centres of various European countries wherein travellers are identified through a face
recognition system installed at various entrance points. On the contrary, AI is also facing some
challenges in terms of credibility, transparency and ethics. Discussion in this direction is
beyond the scope of this article. However, the government and technology providers’
responsibility is to ensure AI-based solutions are explainable, inclusive, non-discriminatory,
available and bias-free (Bhandari et al., 2020).
5.3 Infrastructure providers
In the context of this study, infrastructure is a collective name given to four interdependent
entities, i.e. experts who develop software’s and algorithms, manufacturers who
manufacture specialised hardware devices, network providers that facilitate data exchange
and regulatory body that formulates rules and regulations for the overall technical system.
In general, any information system (IS) is composed of five essential components, i.e. raw
data, manpower, hardware, software and procedures and is treated distinctly from its
working environment (Silver et al.,1995). The simplistic methodology of the interaction of IS
with the environment is the input–process–output mechanism. Paschen et al. (2020)
identified six basic building blocks, i.e. structured data, unstructured data, pre-processes,
main processes, knowledge base and information in developing an AI system, and
demarcated their corresponding roles within the AI system. All these building blocks are
closely concerned with the development and processing of software and algorithms. Roff
(2019) highlights why AI is different from automation and scope of application of AI
depends on the clarity in understanding what the components of AI are. The study stresses
upon four basic components that are vital to developing an AI-based solution for any given
scenario and, i.e. computational power of hardware, domain expert, data and algorithm.
These identified factors cover technical aspects of the AI system. As mentioned above, there
are other factors, including manufacturing, networking and regulatory bodies that drive an
overall system.
Demands for AI-based solutions to solve different socio-economic and political problems
are growing steadily, and this is because of the remarkable development in AI technologies
in the past two decades. Using AI for identifying people is not something which has notbeen
achieved yet. Many private and public sector organisations have implemented AI-based
systems for the identification purpose. Although such systems have shown descent
performance, in practice, there are additional factors that need to be taken into consideration
while developing an AI-enabled identification system for a larger population. Factors such
as parallel computing, scalability, context-aware specialised algorithms, storage capability,
data capture and data processing are some of the parameters that will drive effectiveness
and efficiency of AI-based identity systems. Privacy and security have been identified as
one of the most critical factors in identity systems (Mir et al., 2019,2020b). To support that,
either existing AI techniques have to be tailored if needed, or new techniques need to be
developed that guarantee user-data protection at all levels.
AI-enabled
digital identity
5.4 ID providers
Capability to identify an entity uniquely in the fastest way possible is one of the many pros
of AI technology. Researchers in the past have used AI-based techniques to identify
different types of information from the pool of data. Becerra-Fernadez (2000) used AI for
identifying a particular individual from a knowledge repository based on the expertise of an
individual. The study focuses on the utility of AI in developing People-Finder knowledge
management systems (PFKMS). PFKMS, with the help of AI, identifies the best individual
with the required skill set by querying intellectual capital managed in PFKMS knowledge
base. In another study, Singh (2018) analyses how AI could be used in tracking farmers and
identifying plant diseases through mix mobile and non-mobile devices. Studies such as
Merad et al. (2016) have proposed an AI-based multiple object-tracking solution that
identifies a particular individual from the group. An individual is identified using
behavioural analysis by segmenting target into head, torso and legs and classification of
front and back poses of an individual. Purgason and Hibler (2012) have used pivotal interval
time behavioural biometric data for user identification. The study concludes that the
capability of AI to process and extract unique patterns from biometric data is vital in terms
of identification and classification of individuals. Literature indicates that significant
numbers of identification systems based on AI techniques are primarily using facial
recognition and its variants for identification (Andrejevic and Selwyn, 2019;Beymer and
Poggio, 1995;Okumura et al., 2019,2020).
The best thing about AIeDIS is that there is no need of designated bodies that provide
identity like in traditional identity systems, e.g. financial and educational institutions,
private and public organisations. Identity of a particular individual is the direct outcome of
an AI engine, and that could be dependent on one or more biometric traits of an individual,
e.g. face, voice, eyes,etc. or based on behavioural aspects such as body gestures and walking
style. One of the significant impacts of AI on ID providers is that the AI system itself will
handle most of the existing tasks performed by ID provider and hence makes ID provider
replaceable. AI-based identity system could be free from various human errors that usually
arise because of fatigue. It could also save lots of resources for both individual and
governments in terms of time, cost and efforts.
5.5 Judiciary
ICT, in general, has had a significant impact on the public administration and judicial
system. The literature has supported this wherein the application and implication of digital
technologies in the judicial system have been highlights. For example, Cusatelli and
Giacalone (2015) have studied how ICT is being used in the EU judiciary system. EU has
used ICT to analyse court orders, information exchange, video conferencing between courts,
development of websites, online processing of claims and submission of applications. It not
only has made EU judiciary more efficient but has also empowered common people who can
now see the progress of their pending cases and hence brings transparency in the system.
Overall, it facilitates judges and lawyers in better decision-making. Kiškis and Petrauskas
(2004) have analysed how classification and categorisation problem in the judiciary could be
solved using AI. The authors have chosen Lithuania as a case study, where people do not
have trust in the judiciary because of missing public connect and ambiguity in the working
of the judiciary system. Many researchers are working in understanding the impact of AI on
different sectors and how it will replace humans in the near future. As per McKinsey report,
22% of judicial jobs can be automated (McKinsey, 2017b). In this direction, Sourdin (2018)
studied what roles and responsibilities of judges will be taken over by AI-enabled robots in
coming days. AI technologies facilitate judges in decision-making process, and based on the
JSTPM
complexity of a dispute, it can support or even replace judges in less complex cases. In
another study (Aletras et al., 2016), AI has also been used to predict judicial decisions. The
authors have trained AI system based on the cases processed by the European Court of
Human Rights and proposed a binary classification task that predicts judicial decision with
up to 79% accuracy.
AI also has some peculiar problems that need to be addressed to realise the full potential
of the technology. Levmore and Fagan (2019) have shown that AI systems trained on
historical data may need not be relevant in the future instance. It depends on the coherence
of past and present data variables. It is highly likely that the variables that become relevant
in the future context were missing from the existing AI system and vice versa. Further, the
systems that are formed by combining human and AI deliver superior performance when
compared with only the AI system delivering bail and sentencing decision in the judiciary
system. The main reason that stops authoritiesto hand over full control to AI-based systems
is that AI still operates as “black box”. Researchers have justified why transparency in AI is
a must. The phenomena of justifying why the AI system has taken a particular decision
have been given a specific term called “explainable AI”(xAI). Deeks (2019) studies why xAI
is important in the judiciary and how does it impact in decision-making. AI will bring
efficiency, flexibility and accuracy in various judicial processes including but not limited to
increasing the trial speed, effective judicial resource allocation, interpretation of laws, check
on judicial errors and malpractices and facilitate understanding of judiciary decisions
(Sartor and Branting, 1998).
Any identity system involves collecting, processing and storing personal user data. The
same applies to the AI-enabled identity system. Existing data protection and privacy laws
need to be modified, or new laws need to be enacted such that issues specifictoAI
technology such as credibility and ownership could be addressed. Another important aspect
of an AI-based identity system is to define the legitimate use of user data in online space
when users are availing services from different service providers. To have a robust
regulatory mechanism in place, it is necessary to take regulatory bodies, data protection
authorities, standardising committee and technology providers on-board for effective
regulations.
5.6 Relying parties
Relying parties (RPs) are the entities that provide services to the people by authorising them
using an ID that is provided by a third-party IdP. Identity-related fraud cases are growing
continuously (Pascual et al.,2018) and cyber criminals continuously attack the electronic
communication channel between an IdP and the RP. The fundamental motivation behind
identity theft is to gain access to the services that one is not eligible for. Developing trust in
online communication is a challenging task in the present circumstances when intruder
attacks are becoming common. In the context of electronic communication, Balboni (2004)
analysed the significance of having a liability mechanism in place for certification authority,
and its impact on building trust in online communication. Further certification mechanism
developed by European legislators has been analysed and key points such as promoting
signatory body as RP and removal of reasonable reliance criteria could make certification
mechanism clearer, predictable and easier to define responsibilities. Selection of IdPs by RPs
can have substantial implications for an end-user in terms of data privacy and security.
Vapen et al. (2016) have analysed current third-party identity management environment and
identified what are the most commonly acceptable IdPs by RPs, and how different classes of
RPs vary in selecting their IdPs. Further, it is observed that existing RPs are mostly
websites from North America; however, new websites that are hosted and serving Asian
AI-enabled
digital identity
population and feature in most popular sites are likely to act as replying parties in future.
This bias towards using a few top IdPs makes user’s RP account vulnerable in case user’s
identity credentials are compromised from IdP end.
To solve such problems, many researchers have recommended multi-factor
authentication (MFA) (Bhargav-Spantzel et al.,2007). MFA does solve this problem to a
greater extent, but it also has a downside in terms of user inconvenience and time taken to
log in. In MFA, users are asked to provide additional details in terms of one-time-password,
security question, etc. during login which takes extra time to login and that in turn causes
inconvenience to the end-user (Weir et al.,2009). With AI-enabled authentication, such
problems could be avoided, thereby improving users’overall experience and bringing
efficiency in service delivery of RPs. According to (Columbus, 2019), AI can prevent
negative customers experience by enabling service providers to reduce false-positive cases
and quick approval of the transactions. AI-based identity will enable RPs a safe and secure
mechanism to deliver services to the authenticated users in a timely manner. However, a
transition from current certification-based authentication to authentication of AI-based
identities will require some technological upgradation in the current RP platform to make it
interoperable.
Impact of AI on stakeholders of DI systems and change in roles of each stakeholder are
summarised in Table 7.
6. Discussion
This section highlights the main observations that are synthesised from this conceptual
article. Specific to AIeDIS, we identified five prominent stakeholders, as shown in Figure 4.
It can be concluded that AI can play a vital role in the effectiveness and efficiency of modern
Table 7.
AI vs non-AI identity
systems
Stakeholder Non-AI-enabled identity systems AI-enabled identity systems
Government Public service delivery
Policy development
Regulating overall operations
Customised e-services delivery
Effective policy development based on user
requirements
Identification of marginalised people
User Register by providing personal details
Use unique ID for accessing services
Ensure safe possession of ID
Stress-free registration process
Convenient automatic verification
Scope for user as credentials
Infrastructure Availability of hardware and software
Interconnected devices and databases
Attribute exchange mechanism
Support for large user base and
transactions
Ensure security and privacy of user
credentials
Inclusion of different types of sensors
Inclusion of smart devices for data collection
High-speed connectivity
ID providers Attribute collection
Secure storage of verified user data
Provide unique credentials to users
Approve transactions on behalf of user
Not required
Judiciary Laws to ensure basic human rights are
not violated
Prevent miss use of ID
New laws addressing AI technology concerns
Relying party Provides required set of services to
authorised users
Target-oriented services
Improved access control by constantly
monitoring user interaction patterns
JSTPM
DIS, primarily because of its impact on the identity system’s stakeholders. The reasons why
AI-based identity system is desirable is depicted in Figure 5, highlighting issues with
existing identity system and corresponding solution supported by AI-based identity system.
AI has created much hype among business across sectors, and that isbecause of its potential
to yield massive profits and gain a competitive edge in the highly competitive market space.
AI literature is proliferating, and the majority of the studies are mainly focused on the
technical, ethical and application side of it. Similarly, DI has also gained much attention
recently, and the main driving force could be UNs SDGs-16 i.e. “legal identity to all by 2030”.
The growth in AI technology and the urgency of achieving UNs SDG-16 pave the way for
using disruptive technology for solving societal problems. It has been observed that
disruptive technologies, leaders along with influencers could play a significant role in
achieving SDGs, thereby improving access to services like health care, education, banking,
etc. (Grover et al.,2018,2019). To the best of the authors’knowledge, no study was found in
the literature that has analysed the impact of AI on the stakeholders of the DI system. The
closest has been the studies highlighting the application of biometrics for identification,
authentication and authorisation from a technological perspective (Jacobsen, 2012;Nair,
2018). Besides, some of the studies have attempted to apply AI for different cases of
identification and the same are mentioned in Table 2. It has also been observed that AI-
enabled identity systems increase job performance (Alahmad and Robert, 2020). Hence, in
Figure 4.
Stakeholders of
AIeDIS
Figure 5.
Possible solutions
offered by AI-enabled
identity systems
AI-enabled
digital identity
this conceptual article, we have tried to look into the unexplored dimension of AI technology
and understand how AI could transform DI ecosystem in general and stakeholder
involvement in particular.
According to MacInnis (2011), the contribution of conceptual articles can be made
concerning constructs, relationship or theories, procedures, domains, disciplines and science
and can be classified into four major themes, i.e. envisioning, explicating, relating and
debating. Conceptualisation has been defined as a procedure of abstract thinking that
involves understanding of the mental model of an idea (MacInnis, 2011). This article
contributes to two perspectives, i.e. delineating (explicating) and advocating (debating) at
the procedure level, which includes the nature and range of problems that AIeDIS could
solve.
6.1 Implications to theory
This section highlights the contribution made by this study from a theoretical perspective.
We observed that developing AIeDIS is not only about the technology but the amalgamation
of non-technological factors also. We have tried to extend the DIS literature that primarily
focuses on the issues related to authentication processes.
This conceptual article takes a fresh perspective towards futuristic DIS, which is based
on the impact assessment of AIeDIS on stakeholders. The most important theoretical
implications come in the form of the theoretical model that captures the complex interaction
among various factors of AIeDIS. The theoretical model developed with the help of ST and
IDT may contribute towards a better understanding of the design requirements of AIeDIS.
This is the earliest attempt to use ST, IDT and DT to explain the design and developmental
phenomenon of AIeDIS. Considering the massive budget required in developing DIS, it is
necessary to evaluate each aspect of AIeDIS beforehand so that majority of the risks could
be mitigated at the earliest. For example, in the case of Aadhaar, the total developmental
cost is around 60,000–70,000 crore (McKinsey, 2010;Mir et al., 2020b;Venkatanarayanan,
2018). A proper DIS could save £5bn–£510bn by reducing identity thefts and improving
overall system efficiency (Open Identity Exchange, 2018). This study further strengthens
the understanding regarding the critical factors that could play a vital role in deciding the
success of an AIeDIS.
6.2 Implications to practice
In this article, six main stakeholders of the DI system were identified. These stakeholders
form a complete DI ecosystem. Using AI in DI ecosystem has an impact on all the identified
stakeholders in one way or the other. For example, governments will have to amend existing
policies or devise new ones considering AI capabilities. Governments require a shift in the
policymaking process and adapt to criteria such as customised services; proactive
identification of user requirements that will serve as input to the policy development team;
and distinct governance approach towards the marginalised population. Based on these
transformations, it can be seen that AI could facilitate user empowerment and increased
user participation in government (Carr, 2007;Fischer and Nakakoji, 1992). AI-based
identities will liberate users from the cognitive load, which is the outcome of having to
remember different credentials details for different service providers. Researchers have
identified that because of cognitive load, a user tends toreplicate the same credentials across
service providers and hence makes the system vulnerable to unauthorised access (Horcher
and Tejay, 2009).
Further, AI-based identity has the potential to make identification process frictionless
with the help of real-time behavioural analysis or automatic user verification (e.g. facial
JSTPM
recognition systems in European immigration centres) that in turn could increase user
satisfaction and trust in the overall system. Infrastructure is vital in realising the AI-based
identity system and needs major revamp to make it AI-friendly. Inclusion of sensors for data
collection, uninterrupted high-speed connectivity and support for large user base are what
needs to be ensured for effective AI-based identity system. Upgrading infrastructure would
require a significant chunk of the investment and hence will be a pivotal factor for budget-
constrained countries to implement such a system. Once theproper infrastructure is in place,
it could make IdP redundant as most of the responsibilities of IdPs in traditional identity
system could be handled by infrastructure with least human intervention. In traditional
identity systems, IdP has been the source of many types of errors, including anomalies in
attribute collection, duplicity in assigning credentials and malicious transaction approval
(Henneman et al., 2008). Having these tasks as part of infrastructure could result in a better
and efficient identity system as all these activities will be handled by machines and sensors
with almost zero human intervention and hence reduces chances of human-originated
errors. There is a continuous debate going on among lawyers, researchers, academicians,
NGOs, technologists and governments regarding AI and its ownership. Questions such as
who should be responsible for the consequences of AI-based system; and is it the
technologist who developed sensors or programmer who wrote algorithms or government
that allowed commercial use of such technology are still actively debated (Robles Carrillo,
2020). AI-based identity systems pose novel challenges for the judiciary that needs to be
addressed clearly through the rule of law. Laws that could ensure human rights are not
violated by AI-enabled identity system need to be approved before commercialising it. The
biggest challenge for judiciary is regarding the geo-location of the user database that
contains all the critical credential details about the user and its access rights (Selby, 2017).
Judiciary and AI in itself is a separate area, and many researchers are actively working in
this direction. Detailed analysis in this direction is outside the scope of this paper, but some
researchers such as Barredo Arrieta et al. (2020),Kaplan and Haenlein (2020) and Wang et al.
(2015) have addressed some of the recent challenges that AI-based technologies pose. AI-
enabled identity systems will open new revenue channels for RPs. Having a better
understanding of the user base will enable RPs to serve its customers better by
recommending tailored services based on users’needs and preferences. Further, better
access control mechanism will avoid malicious access requests and, in turn, will lead to
service availability that could improve the quality of service and increase in user trust.
The article highlights the details of AI-based identity and its impact on the stakeholders.
It is observed that all stakeholders could be impacted and could bring efficiency in the
working of stakeholders. For example, governments will have a better understanding about
its citizens and their demands in advance, which will enable them to be proactive in their
approach to avoid any uprising that could be possible in the near future (Kar, 2020;Mohan
and Kar, 2017). If done right, it could make both government and citizens empowered and
will eventually result in increased trust in government. Inclusion of intelligent devices as
part of infrastructure will increase the capability of the system and can handle tasks that
were treated by separate entities such as IdPs in this case. Most of the activities of the IdPs
will be conveniently handled as one sub-activity of infrastructure, resulting in more nuanced
access control mechanisms (Phiri and Agbinya, 2006). For judiciary, AI-based identity
system has opened an entirely new dimension of justice –justice for and from autonomous
systems. In one way, AI-based systems will facilitate judiciary to make better decisions by
avoiding human bias in sentencing processes and, on the other hand, requires the judiciary
to devise new laws that addressthe challenges posed by AI-based technology.
AI-enabled
digital identity
AI brings multi-fold benefits for RPs in terms of fool-proof access control, customised
service delivery, better knowledge about customers and robustness in service delivery.
Apart from contributing to the DIS literature and practice, the research also contributes to
the methodology, as it is for the first time that three prominent theories, i.e. ST, DT and IDT,
were used to study the design aspects of DIS in general and AIeDIS in particular. The proposed
AIeDIS DT could be used as an initial framework for developing new-age innovation-based
identity systems in future. It could enable concerned authorities to have an idea about the type
of issues innovations can address and at what cost. Needless to mention that the experience
gained may be useful in analysing other large-scale innovation-driven studies in future.
Finally, based on this study’sfindings, integrating AI with DI systems is advocated
considering the range of benefits it might yield for all the concerned stakeholders of the
system. Based on the authors’perception, a comparative analysis between AI-based identity
system, traditional and non-AI DI system is shown in Table 8, highlighting in what aspects
AI enhances existing identity systems. Hence, it is recommended that the inclusion of AI
technology for identification should be commercialised with mandatory regulations in place
such that the risks associated with current identity systems such as identity theft, duplicity,
unauthorised access, reply attacks and bias could be addressed in addition to opening a new
horizon of intelligent service-delivery mechanisms.
6.3 Limitations and future research
The study is based on secondary data and has considered DI stakeholders from a generic
perspective. It will be further benefited by incorporating primary data for empirical analysis
that focuses on the impact of AI on each stakeholder’s sub-processes and testing the
hypothesis identified in Table 4. Directions for future research are to validate this study’s
results by incorporating expert opinion for analysis and developing a prototype of AIeDIS
for better visualisation.
7. Conclusion
In this conceptual article, we identified six significant stakeholders of general-purpose DIS
and five stakeholders of AIeDIS from the secondary data sources that primarily included
research articles, official reports, white papers and publically available interviews of domain
experts. The findings depict that AIeDIS will not only solve some crucial problems such as
identity theft, unauthorised access and credential misuse but will also open a possibility of
new ways to empower all stakeholders in one way or the other. The inclusion of AI in DIS
makes it easy to determine true identity, hard to circumvent and forge and promptly process
a large number of requests.
Table 8.
Comparative
analysis of different
variants of identity
systems
Factor Traditional identity Digital identity AI-enabled digital identity
Type Functional Functional/foundational Foundational
Application Service-specific Diverse Across sectors
Suitable for online No Yes Yes
Implementation Easy Moderate Complex
Security Low Medium High
Interoperability Low Medium High
Developmental cost Low High High
On-boarding cost High Low Low
Duplicates High Low Very low
Scalability Low Medium High
JSTPM
From the theoretical perspective, the study explores ST and IDT’s utility in the design
and development process outlined in DT. To the best of the authors’knowledge, this is the
first instance wherein three prominent theories, i.e. DT, ST and IDT, have been used to
highlight the complex interaction mechanism among the stakeholders of AIeDIS. Apart
from all the potential advantages of AIeDIS, there are still some issues such as explanation
and ownership of decisions that are yet to be addressed and have been the main pain points
for the authorities to call on the commercialisation of AIeDIS for large scale. This study acts
as an aid for the decision makers to decide whether to include AI as a core component in the
DIS or not primarily for the governments in the initial phase of DIS development or are
attempting to retrofit it in the existing functional DIS for better utility.
References
Abedin, B., Abedin, B., Khoei, T.T. and Ghapanchi, A.R. (2013), “A review of critical factors for
communicating with customers on social networking sites”,The International Technology
Management Review, Vol. 3 No. 4, p. 208.
Agbinya, J.I. (2019), 16 Digital Identity Management System Using Artificial Neural Networks (Applied
Da), River Publishers.
Agrawal, A.K., Gans, J. and Goldfarb, A. (2018), “Economics of artificial intelligence”, available at:
https://conference.nber.org/conferences/2017/AIf17/summary.html
Alahmad, R. and Robert, L. (2020), “Artificial intelligence (AI) and IT identity: antecedents identifying
with AI applications”,Proceedings of the 26th Americas Conference on Information Systems,
pp. 13-15.
Albiman, M.M. and Sulong, Z. (2016), “The role of ICT use to the economic growth in Sub Saharan
African region (SSA)”,Journal of Science and Technology Policy Management, Vol. 7 No. 3,
pp. 306-329.
Aletras, N., Tsarapatsanis, D., Preotiuc-Pietro, D. and Lampos, V. (2016), “Predicting judicial decisions
of the European court of human rights: a natural language processing perspective”,PeerJ
Computer Science, Vol. 2, p. e93.
Al-Khouri, A.M. (2014), “Digital identity: transforming GCC economies”,Innovation, Vol. 16 No. 2,
pp. 184-194.
Allison, A., Currall, J., Moss, M. and Stuart, S. (2005), “Digital identity matters”,Journal of the American
Society for Information Science and Technology, Vol. 56 No. 4, pp. 364-372.
Amato, F., L
opez, A., Peña-Méndez, E.M., Vanõhara, P., Hampl, A. and Havel, J. (2013), “Artificial
neural networks in medical diagnosis”,Journal of Applied Biomedicine, Vol. 11 No. 2, pp. 47-58.
Andrejevic, M. and Selwyn, N. (2019), “Facial recognition technology in schools: critical questions and
concerns”,Learning, Media and Technology, Vol. 45 No. 2, pp. 1-14.
Androutsopoulou, A., Karacapilidis, N., Loukis, E. and Charalabidis, Y. (2019), “Transforming the
communication between citizens and government through AI-guided chatbots”,Government
Information Quarterly, Vol. 36 No. 2, pp. 358-367.
Ashaye, O.R. and Irani, Z. (2019), “The role of stakeholders in the effective use of e-government resources
in public services”,International Journal of Information Management, Vol. 49, pp. 253-270.
Atick, J. (2016), “Digital identity: the essential guide”, available at: www.id4africa.com/prev/%0Aimg/
Digital_Identity_The_Essential_Guide.pdf
Ayoub, K. and Payne, K. (2016), “Strategy in the age of artificial intelligence”,Journal of Strategic
Studies, Vol. 39 Nos 5/6, pp. 793-819.
Balboni, P. (2004), “Liability of certification service providers towards relying parties and the need for a
clear system to enhance the level of trust in electronic communication”,Information and
Communications Technology Law, Vol. 13 No. 3,pp. 211-242.
AI-enabled
digital identity
BankWorld (2016), “World development report 2016: digital dividends”, The World Bank, available at:
https://doi.org/10.1596/978-1-4648-0671-1
Barredo Arrieta, A., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., García, S.,
Gil-L
opez, S., Molina, D., Benjamins, R. and Herrera, F. (2020), “Explainable artificial intelligence
(XAI): concepts, taxonomies, opportunities and challenges toward responsible AI”,Information
Fusion, Vol. 58, pp. 82-115.
Bates, J. (1992), “Virtual reality, art, and entertainment”,Presence: Teleoperators and Virtual
Environments, Vol. 1 No. 1, pp. 133-138.
Becerra-Fernadez, I. (2000), “The role of artificial intelligence technologies in the implementation of
people-finder knowledge management systems”,Knowledge-Based Systems, Vol. 13 No. 3,
pp. 315-320, doi:10.1016/S0950-7051(00)00091-5.
Beduschi, A. Cinnamon, J. Langford, J. Luo, C. and Owen, D. (2017), “Building digital identities”,
Economic and Social Research Council, Vol. 40.
Belk, R.W. (2013), “Extended self in a digital world”,Journal of Consumer Research, Vol. 40 No. 3, pp. 477-500.
Beymer, D. and Poggio, O. (1995), “Face recognition from one example view”,Proceedings of IEEE
International Conference on Computer Vision,IEEE, pp. 500-507.
Bhandari, V., Trikanad, S. and Sinha, A. (2020), “Governing ID: a framework for evaluation of digital
identity. The Centre for internet and society”, available at: https://digitalid.design/docs/
CIS_DigitalID_EvaluationFrameworkDraft02_2020.01.pdf
Bhargav-Spantzel, A., Squicciarini, A. and Bertino, E. (2007), “Privacy preserving multi-factor
authentication with biometrics”,Journal of Computer Security, Vol.15 No. 5, pp. 529-560.
Bhattacherjee (2020), Social Science Research, Principles, Methods, and Practices, LibreTexts, FL.
Bolton, R.J. and Hand, D.J. (2002), “Statistical fraud detection: a review”,Statistical Science, Vol. 17 No. 3,
pp. 235-249.
Bowman, N. and Banks, J. (2019), “Social and entertainment gratifications of videogame play
comparing robot, AI, and human partners”,2019 28th IEEE International Conference on Robot
and Human Interactive Communication (RO-MAN),New Delhi, pp. 1-6.
Brynjolfsson, E. and Mitchell, T. (2017), “What can machine learning do? Workforce implications”,
Science, Vol. 358 No. 6370, pp. 1530-1534.
Buhalis, D., Harwood, T., Bogicevic, V., Viglia, G., Beldona, S. and Hofacker, C. (2019), “Technological
disruptions in services: lessons from tourism and hospitality”,Journal of Service Management,
Vol. 30 No. 4, pp. 484-506.
Cameron, K. (2005), “The laws of identity”, available at: http://myinstantid.com/laws.pdf
Camp, L.J. (2004), “Digital identity”,IEEE Technology and Society Magazine, Vol. 23 No. 3, pp. 34-41.
Carr, S. (2007), “Participation, power, conflict and change: theorizing dynamics of service user participation
in the social care system of England and Wales”,Critical Social Policy, Vol. 27 No. 2, pp. 266-276.
Chatterjee, S. and Kar, A.K. (2018), “Effects of successful adoption of information technology enabled
services in proposed smart cities of India: from user experience perspective”,Journal of Science
and Technology Policy Management, Vol. 9 No. 2, pp. 189-209.
Clark, J. Dahan, M. Desai, V. Ienco, M. De Labriolle, S. Pellestor, J.-P. …Varuhaki, Y. (2016), “Digital
identity: towards shared principles for public and private sector cooperation, (July)”, pp. 1-44,
available at: www.worldbankgroup.org/id4d
Cole, D. (1991), “Artificial intelligence and personal identity”,Synthese, Vol. 88 No.3, pp. 399-417.
Columbus, L. (2019), “Top 9 ways artificial intelligence prevents fraud”, available at: www.forbes.com/
sites/louiscolumbus/2019/07/09/top-9-ways-artificial-intelligence-prevents-fraud/#3748f57f14b4
Cusatelli, C. and Giacalone, M. (2015), “ICT use by judiciary systems in European union”,Journal of
Applied Quantitative Methods, Vol. 10 No. 2, pp. 1-6.
JSTPM
de Sousa, W.G., de Melo, E.R.P., Bermejo, P.H.D.S., Farias, R.A.S. and Gomes, A.O. (2019), “How and
where is artificial intelligence in the public sector going? A literature review and research
agenda”,Government Information Quarterly, Vol. 36 No. 4, p. 101392.
de Vries, K. (2010), “Identity, profiling algorithms and a world of ambient intelligence”,Ethics and
Information Technology, Vol. 12 No. 1, pp. 71-85.
Deeks, A. (2019), “The judicial demand for explainable artificial intelligence”, (August 1, 2019). 119
Colum. L. Rev. (2019 Forthcoming), Virginia Public Law and Legal Theory Research Paper
No. 2019-51, available at: https://ssrn.com/abstract=3440723
Desai, V.T., Diofasi, A. and Lu, J. (2018), “The global identification challenge: who are the 1 billion
people without proof of identity?”, available at: https://blogs.worldbank.org/voices/global-
identification-challenge-who-are-1-billion-people-without-proof-identity
Dixon, P. (2017), “A failure to do no harm –India’s Aadhaar biometric ID program and its inability to protect
privacy in relation to measures in Europe and the US”,Health and Technology, Vol. 7 No. 4, pp. 539-567.
Duijst, D. (2017), Can we Improve the User Experience of Chatbots withPersonalisation?, University of
Amsterdam.
Dunn, W. (2017), Public Policy Analysis: An Integrated Approach, 6th ed., Routledge, New York, NY.
Dwivedi, Y.K., Hughes, L., Ismagilova, E., Aarts, G., Coombs, C., Crick, T., Duan, Y., Dwivedi, R.,
Edwards, J., Eirug, A. and Williams, M.D. (2019), “Artificial intelligence (AI): multidisciplinary
perspectives on emerging challenges, opportunities, and agenda for research, practice and
policy”,International Journal of Information Management, Vol. 57, p. 101994.
Ellemers, N., Spears, R. and Doosje, B. (2002), “Self and social identity”,Annual Review of Psychology,
Vol. 53 No. 1, pp. 161-186.
Ema,A.,Akiya,N.,Osawa,H.,Hattori,H.,Oie,S.,Ichise,R.,Kanzaki,N.,Kukita,M.,Saijo,R.,Takushi,O.
and Yashiro, Y. (2016), “Future relations between humans and artificial intelligence: a stakeholder
opinionsurveyinJapan”,IEEE Technology and Society Magazine, Vol. 35 No. 4, pp. 68-75.
Fischer, G. and Nakakoji, K. (1992), “Beyond the macho approach of artificial intelligence: empower
human designers –do not replace them”,Knowledge-Based Systems, Vol. 5 No. 1, pp. 15-30.
Flasi
nski, M. (2016), Introduction to Artificial Intelligence, Springer.
Freeman, R.E. (2015), “Stakeholder theory”,Wiley Encyclopedia of Management, John Wiley and Sons,
Chichester, pp. 1-6, doi: 10.1002/9781118785317.weom020179.
Goldkuhl, G. (2004), “Design theories in information systems –a need for Multi-Grounding”,Journal of
Information Technology Theory and Application (JITTA), Vol. 6 No. 2, pp. 59-72, available at:
https://aisel.aisnet.org/cgi/viewcontent.cgi?article=1127&context=jitta
Goode, J. (2010), “The digital identity divide: how technology knowledge impacts college students”,
New Media and Society, Vol. 12 No. 3, pp. 497-513.
Goodell, G. and Aste, T. (2019), “A decentralized digital identity architecture”,Frontiers in Blockchain,Vol.2.
Gregor, S. and Jones, D. (2007), “The anatomy of a design theory”,Journal of the Association for
Information Systems, Vol. 8 No. 5, pp. 312-335, available at: http://citeseerx.ist.psu.edu/viewdoc/
download?doi=10.1.1.232.743&rep=rep1&type=pdf
Grover, P., Kar, A.K. and Davies, G. (2018), “‘Technology enabled health’–insights from twitter analytics with
a socio-technical perspective”,International Journal of Information Management, Vol. 43, pp. 85-97.
Grover, P., Kar, A.K. and Ilavarasan, P.V. (2019), “Impact of corporate social responsibility on
reputation–insights from tweets on sustainable development goals by CEOs”,International
Journal of Information Management, Vol. 48, pp. 39-52.
Hatchuel, A., Le Masson, P., Weil, B., Agogué, M., Kazakçi, A. and Hooge, S. (2016), “Multiple forms of
applications and impacts of a design theory -ten years of industrial applications of C-K theory”,Impact
of Design Research on Industrial Practice -Tools, Technology, and Training, Vol. 1 No. 1, pp. 189-208.
AI-enabled
digital identity
Helbing, D., Frey, B.S., Gigerenzer, G., Hafen, E., Hagner, M., Hofstetter, Y., Van Den Hoven, J.,
Zicari, R.V. and Zwitter, A. (2019), “Will democracy survive big data and artificial intelligence?”,
Towards Digital Enlightenment, Springer International Publishing, Cham, pp. 73-98,
doi: 10.1007/978-3-319-90869-4_7.
Henneman, P.L., Fisher, D.L., Henneman, E.A., Pham, T.A., Mei, Y.Y., Talati, R., Nathanson, B.H. and
Roche, J. (2008), “Providers do not verify patient identity during computer order entry”,
Academic Emergency Medicine, Vol. 15 No.7, pp. 641-648.
Horcher, A.-M. and Tejay, G.P. (2009), “Building a better password: the role of cognitive load in
information security training”,2009 IEEE International Conference on Intelligence and Security
Informatics, IEEE, pp. 113-118.
Horowitz, M.C., Allen, G.C., Saravalle, E., Cho, A., Frederick, K. and Scharre, P. (2018), Artificial
Intelligence and International Security, Center for a New American Security.
ITU-T FG-DFS (2017), “Review of national identity programs”,Itu-T Fg-Dfs.
Jacobsen, E.K.U. (2012), “Unique identification: inclusion and surveillance in the Indian biometric
assemblage”, available at: https://doi.org/10.1177/0967010612458336
Janssen, M. and Estevez, E. (2013), “Lean government and platform-based governance –doing more
with less”,Government Information Quarterly, Vol. 30, pp. S1-S8.
Jung, J. and Mittal, V. (2020), “Political identity and the consumer journey: a research review”,Journal of
Retailing, Vol. 96 No. 1,pp. 55-73.
Kaplan, A. and Haenlein, M. (2020), “Rulers of the world, unite! The challenges and opportunities of
artificial intelligence”,Business Horizons, Vol. 63 No. 1, pp. 37-50.
Kar, A.K. (2016), “Bio inspired computing –a review of algorithms and scope of applications”,Expert
Systems with Applications, Vol. 59, pp. 20-32.
Kar, A.K. (2020), “What affects usage satisfaction in mobile payments? Modelling user generated
content to develop the ‘digital service usage satisfaction model’”,Information Systems Frontiers,
doi: 10.1007/s10796-020-10045-0.
Kiškis, M. and Petrauskas, R. (2004), “ICT adoption in the judiciary: classifying of judicial information”,
International Review of Law, Computers and Technology, Vol. 18 No.1, pp. 37-45.
Knight, A. and Saxby, S. (2014), “Identity crisis: global challenges of identity protection in a networked
world”,Computer Law and Security Review, Vol. 30No. 6, pp. 617-632.
Ku, C.-H. and Leroy, G. (2014), “A decision support system: automated crime report analysis and
classification for e-government”,Government Information Quarterly, Vol. 31 No. 4, pp. 534-544.
Lalmuanawma, S., Hussain, J. and Chhakchhuak, L. (2020), “Applications of machine learning and
artificial intelligence for covid-19 (SARS-CoV-2) pandemic: a review”,Chaos, Solitons and
Fractals, Vol. 139, p. 110059.
Larson, C. (2018), “China’s AI imperative”,Science, Vol. 359 No. 6376, pp. 628-630, doi:10.1126/
science.359.6376.628.
Laurent, M. and Bouzefrane, S. (2015), “Digital identity management. ISTE press and Elsevier”,
available at: https://doi.org/10.1007/978-3-319-08231-8
Lauterbach, A. (2019), “Artificial intelligence and policy: quo vadis?”,Digital Policy, Regulation and
Governance, Vol. 21No. 3, pp. 238-263.
Levmore, S. and Fagan, F. (2019), “The impact of artificial intelligence on rules, standards, and judicial
discretion”,SSRN Electronic Journal, Vol. 93 No. 1, doi: 10.2139/ssrn.3362563.
Li, J.(J.)., Bonn, M.A. and Ye, B.H. (2019), “Hotel employee’sartificial intelligence and robotics
awareness and its impact on turnover intention: the moderating roles of perceived
organizational support and competitive psychological climate”,Tourism Management,
Vol. 73, pp. 172-181.
JSTPM
Liang, Y., Samtani, S., Guo, B. and Yu, Z. (2020), “Behavioral biometrics for continuous authentication
in the internet-of-things era: an artificial intelligence perspective”,IEEE Internet of Things
Journal, Vol. 7 No. 9, pp. 9128-9143.
Lim, S., Saldanha, T.J.V., Malladi, S. and Melville, N.P. (2013), “Theories used in information system
research: insights from complex network analysis”,Journal of Information Technology Theory
and Application, Vol. 14 No. 2, pp. 5-46.
Liu, Q., Diao, L. and Tu, G. (2010), “The application of artificial intelligence in mobile learning”,
2010 International Conference on System Science, Engineering Design and Manufacturing
Informatization,IEEE, pp. 80-83.
McKinsey (2010), “Inclusive growth and financial security”, available at: http://ccmrm.org/wp-content/
uploads/2015/05/McKinsey-2010-inclusive-growth-report.pdf
McKinsey (2017a), “Artificial intelligencethe next digital frontier?”, Available at: www.mckinsey.
com//media/McKinsey/Industries/AdvancedElectronics/OurInsights/Howartificialintelligencecan
deliverrealvaluetocompanies/MGI-Artificial-Intelligence-Discussion-paper.ashx
McKinsey (2017b), “Jobs lost, jobs gained: workforce transitions in a time of automation”, available at:
www.mckinsey.com//media/mckinsey/featuredinsights/FutureofOrganizations/Whatthefutureof
workwillmeanforjobsskillsandwages/MGI-Jobs-Lost-Jobs-Gained-Report-December-6-2017.ashx
McKinsey (2019), “Digital identification: a key to inclusive growth”, available at: www.mckinsey.
com//media/mckinsey/featuredinsights/innovation/thevalueofdigitalidfortheglobaleconomy
andsociety/mgi-digital-identification-a-key-to-inclusive-growth.ashx
Mcwaters, R.J. (2016), “A blueprint for digital identity the role of financial institutions in building
digital identity”,World Economic Forum, Future of Financial Services Series, (August),
pp. 1-108, available at: www3.weforum.org/docs/WEF_A_Blueprint_for_Digital_Identity.pdf
MacInnis, D.J. (2011), “A framework for conceptual contributions in marketing”,Journal of Marketing,
Vol. 75 No. 4, pp. 136-154.
Merad, D., Aziz, K.-E., Iguernaissi, R., Fertil, B. and Drap, P. (2016), “Tracking multiple persons under
partial and global occlusions: application to customers’behavior analysis”,Pattern Recognition
Letters, Vol. 81, pp. 11-20.
Milano, M., O’Sullivan, B. and Gavanelli, M. (2014), “Sustainable policy making: a strategic challenge
for artificial intelligence”,AI Magazine, Vol. 35 No. 3, p. 22.
Miller, H. and Stirling, R. (2019), “Government artificial intelligence readiness index 2019”, available at:
www.oxfordinsights.com/ai-readiness2019
Mir, U.B., Kar, A.K. and Gupta, M.P. (2020a), “Digital identity evaluation framework for social welfare”,
IFIP Advances in Information and Communication Technology, Vol. 617, pp. 401-414,
doi: 10.1007/978-3-030-64849-7_36.
Mir, U.B., Kar, A.K., Gupta, M.P. and Sharma, R.S. (2019), “Prioritizing digital identity goals –the case
study of Aadhaar in India”,Digital Transformation for a Sustainable Society in the 21st Century,
pp. 489-501, doi: 10.1007/978-3-030-29374-1_40.
Mir, U.B., Sharma, S., Kar, A.K. and Gupta, M.P. (2020c), “Critical success factors for integrating
artificial intelligence and robotics”,Digital Policy, Regulation and Governance, Vol. 22 No. 4.
Mir, U.B., Kar, A.K., Dwivedi, Y.K., Gupta, M.P. and Sharma, R.S. (2020b), “Realizing digital identity in
government: prioritizing design and implementation objectives for Aadhaar in India”,
Government Information Quarterly, Vol. 37 No. 2, p. 101442.
Mishra, A. and Mishra, D. (2013), “Applications of stakeholder theory in information systems and
technology”,Engineering Economics, Vol. 24 No. 3.
Moavenzadeh, J. and de Maar, L. (2018), “The known traveller: unlocking the potential of digital
identity for secure and seamless travel”,World Economic Forum, (January), available at: www3.
weforum.org/docs/WEF_The_Known_Traveller_Digital_Identity_Concept.pdf
AI-enabled
digital identity
Mohan, R. and Kar, A.K. (2017), “#demonetization and its impact on the Indian economy –insights from
social media analytics”, pp. 363-374, available at: https://doi.org/10.1007/978-3-319-68557-1_32
Motu’apuaka, M., Whitlock, E., Kato, E., Uhl, S., Belinson, S., Chang, C., Hoomans, T., Meltzer, D.O.,
Noorani, H., Robinson, K.A. and Cottrell, E. (2015), “Defining the benefits and challenges of
stakeholder engagement in systematic reviews”,Comparative Effectiveness Research, Vol. 13,
doi: 10.2147/CER.S69605.
Nair, V. (2018), “An eye for an I: recording biometrics and reconsidering identity in postcolonial India”,
Contemporary South Asia, Vol. 26 No. 2, pp. 143-156.
Newman,G.R.andMcNally,M.M.(2005),“Identity theft literature review”, No. 210459, United States of America,
available at: https://www.ojp.gov/ncjrs/virtual-library/abstracts/identity-theft-literature-review
OCED (2011), Digital Identity Management: Enabling Innovation and Trust in The Internet Economy,
Organisation for Economic Co-operation and Development (OECD), United States.
Okumura, A., Komeiji, S., Sakaguchi, M., Tabuchi, M. and Hattori, H. (2019), “Identity verification using face
recognition for artificial-intelligence electronic forms with speech interaction”,HCI for Cybersecurity,
Privacy and Trust, pp. 52-66, available at: https://doi.org/10.1007/978-3-030-22351-9_4
Okumura, A., Handa, S., Hoshino, T., Tokunaga, N. and Kanda, M. (2020), “Identity verification using
face recognition improved by managing check-in behavior of event attendees”,Advances in
Artificial Intelligence, pp. 291-304, available at: https://doi.org/10.1007/978-3-030-39878-1_26
Olson, E.T. (2015), “Personal identity”, available at: https://plato.stanford.edu/entries/identity-personal/
(accessed 21 April 2019).
Open Identity Exchange (2018), Digital Identity in The UK : The Cost of Doing Nothing, United Kingdom.
Pan, Y. (2016), “Heading toward artificial intelligence 2.0”,Engineering, Vol. 2 No. 4, pp. 409-413.
Pantano, E. and Pizzi, G. (2020), “Forecasting artificial intelligence on online customer assistance: evidence
from chatbot patents analysis”,Journal of Retailing and Consumer Services, Vol. 55, p. 102096.
Paschen, U., Pitt, C. and Kietzmann, J. (2020), “Artificial intelligence: building blocks and an innovation
typology”,Business Horizons, Vol. 63 No. 2, pp. 147-155.
Pascual, A., Marchini, K. and Miller, S. (2018), “2018 Identity fraud: fraud enters a new era of
complexity”, available at: www.javelinstrategy.com/coverage-area/2018-identity-fraud-fraud-
enters-new-era-complexity
Passonneau, R.J., McNamara, D., Muresan, S. and Perin, D. (2017), “Preface: special issue on
multidisciplinary approaches to AI and education for reading and writing”,International
Journal of Artificial Intelligence in Education, Vol. 27 No. 4, pp. 665-670.
Phiri, J. and Agbinya, J.I. (2006), “Modelling and information fusion in digital identity management
systems”,International Conference on Networking, International Conference on Systems
and International Conference on Mobile Communications and Learning Technologies
(ICNICONSMCL’06), pp. 181-181.
Phiri, J., Zhao, T.-J. and Agbinya, J.I. (2011a), “Biometrics, device metrics and pseudo metrics in a multifactor
authentication with artificial intelligence”,Proceedings of the 6th International Conference on
Broadband Communications and Biomedical Applications,IEEE,Melbourne,pp.1-6.
Phiri, J., Zhao, T.-J., Zhu, C.H. and Mbale, J. (2011b), “Using artificial intelligence techniques to
implement a multifactor authentication system”,International Journal of Computational
Intelligence Systems, Vol. 4 No. 4, pp. 420-430.
Power, D.J. (2016), “‘Big brother’can watch us”,Journal of Decision Systems, Vol. 25 No. sup1, pp. 578-588.
Purgason, B. and Hibler, D. (2012), “Security through behavioral biometrics and artificial intelligence”,
Procedia Computer Science, Vol. 12, pp. 398-403.
Rahman, M.S. and Ko, M. (2013), “Toward systematic identification of stakeholders for healthcare
information systems: a feature-based method”,Proceedings of the Nineteenth Americas
Conference on Information Systems,Chicago, IL.
JSTPM
Rai, A., Constantinides, P. and Sarker, S. (2018), “Editor’s comments: next-generation digital platforms:
toward human–AI hybrids”,MIS Quarterly, Vol. 43 No. 1, pp. 3-9.
Rifkin, J. (2001), The Age of Access: The New Culture of Hypercapitalism Where All of Life is a
Paid-For Experience, Putnam Publishing Group, available at: https://dl.acm.org/citation.
cfm?id=518530
Rissland, E.L., Ashley, K.D. and Loui, R.P. (2003), “AI and law: a fruitful synergy”,Artificial Intelligence,
Vol. 150 Nos 1/2, pp. 1-15.
Robles Carrillo, M. (2020), “Artificial intelligence: from ethics to law”,Telecommunications Policy,
Vol. 44 No. 6, p. 101937.
Roff, H.M. (2019), “Artificial intelligence: power to the people”,Ethics and International Affairs, Vol. 33
No. 2, pp. 127-140.
Ronald, A. Elizabeth, S.B. Noopur, S. and Neil, S.B. (2017), “State of Aadhaar report 2016-17”, India.
Rowley, J. (2011), “e-Government stakeholders –who are they and what do they want?”,International
Journal of Information Management, Vol. 31 No. 1, pp. 53-62.
Ryman-Tubb, N.F., Krause, P. and Garn, W. (2018), “How artificial intelligence and machine learning
research impacts payment card fraud detection: a survey and industrybenchmark”,Engineering
Applications of Artificial Intelligence, Vol. 76,pp. 130-157.
Santeli, J.T. and Gerdon, S. (2019), “5 Challenges for government adoption of AI”, available at: www.
weforum.org/agenda/2019/08/artificial-intelligence-government-public-sector/
Sartor, G. and Branting, L. (1998), Judicial Applications of Artificial Intelligence, in Sartor, G. and
Branting, K. (Eds), Springer Netherlands, Dordrecht, doi: 10.1007/978-94-015-9010-5.
Schoemaker, E., Baslan, D., Pon, B. and Dell, N. (2020), “Identity at the margins:data justice and refugee
experiences with digital identity systems in Lebanon, Jordan, and Uganda”,Information
Technology for Development, Vol. 27 No. 1, pp. 13-36.
Selby, J. (2017), “Data localization laws: trade barriers or legitimate responses to cybersecurity risks, or
both?”,International Journal of Law and Information Technology, Vol. 25 No. 3, pp. 213-232.
Shank, D.B., Graves, C., Gott, A., Gamez, P. and Rodriguez, S. (2019), “Feeling our way to machine
minds: people’s emotions when perceiving mind in artificial intelligence”,Computers in Human
Behavior, Vol. 98, pp. 256-266.
Sharma, R.S. (2016), “UIDAI’s public policy innovations”, National Institute of Public Finance and
Policy, No. 176, pp. 1-19.
Sharma, G.D., Yadav, A. and Chopra, R. (2020), “Artificial intelligence and effective governance: a
review, critique and research agenda”,Sustainable Futures, Vol. 2, p. 100004.
Silver, M.S., Markus, M.L. and Beath, C.M. (1995), “The information technology interaction model: a
foundation for the MBA core course”,MIS Quarterly, Vol. 19 No. 3, p. 361.
Simon, J.P. (2019), “Artificial intelligence: scope, players, markets and geography”,Digital Policy,
Regulation and Governance, Vol. 21 No. 3, pp. 208-237.
Singh, K.K. (2018), “An artificial intelligence and cloud based collaborative platform for plant disease
identification, tracking and forecasting for farmers”,2018 IEEE International Conference on
Cloud Computing in Emerging Markets (CCEM), IEEE, pp. 49-56.
Sourdin, T. (2018), “Judge v. Robot: artificial intelligence and judicial Decision-Making”,UNSW Law
Journal, Vol. 4 No. 4, pp. 1114-1133.
Singh, H., Kar, A.K. and Ilavarsana, P.V. (2017), “Performance assessment of e-government projects:
a multi-construct, multi-stakeholder perspective”,Proceedings of the 10th International
Conference on Theory and Practice of Electronic Governance, pp. 558-559.
Strich, F., Mayer, A.-S. and Fiedler, M. (2021), “What do I do in a world of artificial intelligence? Investigating
the impact of substitutive decision-making AI systems on employees’professional role identity”,
Journal of the Association for Information Systems,Vol.22No.2,doi:10.17705/1jais.00663.
AI-enabled
digital identity
Sun, T.Q. and Medaglia, R. (2019), “Mapping the challenges of artificial intelligence in the public sector:
evidence from public healthcare”,Government Information Quarterly, Vol. 36 No. 2, pp. 368-383.
Tajitsu, N. (2016), “Toyota to build artificial intelligence-based driving systems in five years”, available
at: https://in.reuters.com/article/us-toyota-ai/toyota-to-build-artificial-intelligence-based-driving-
systems-in-five-years-idINKCN0Z60BE
Toll,D.,Lindgren,I.,Melin,U.andMadsen,C.Ø.(2020),“Values, benefits, considerations and risks of AI in
government”,JeDEM - EJournal of EDemocracy and Open Government, Vol. 12 No. 1, pp. 40-60.
Van De Kaa, G., Janssen, M. and Rezaei, J. (2018), “Technological forecasting and social change
standards battles for business-to-government data exchange: identifying success factors for
standard dominance using the best worst method”,Technological Forecasting and Social
Change, Vol. 137, pp. 182-189.
Vapen, A., Carlsson, N., Mahanti, A. and Shahmehri, N. (2016), “A look at the third-party identity
management landscape”,IEEE Internet Computing, Vol. 20 No. 2, pp. 18-25.
Venkatanarayanan, A. (2018), “Aadhaar enrolment costs”, available at: https://medium.com/karana/
aadhaar-enrolment-costs-bc17f0d30018 (accessed 4 July 2019).
Verma, C., Tarawneh, A.S., Illes, Z., Stoffova, V. and Singh, M. (2019), “National identity predictive
models for the real time prediction of European school’s students: preliminary results”,
2019 International Conference on Automation, Computational and Technology Management
(ICACTM), IEEE, pp. 418-423.
W.Sadek, A. (2007), “Artificial intelligence applications in transportation”,Artificial Intelligence in
Transportation: Information for Application, pp. 1-6.
Walls, J.G., Widmeyer, G.R. and El Sawy, O.A. (1992), “Building an information system design theory
for vigilant EIS”,Information Systems Research, Vol. 3 No. 1, pp. 36-59.
Wang,X.,Li,X.andLeung,V.C.M.(2015),“Artificial intelligence-based techniques for emerging heterogeneous
network: state of the arts, opportunities, and challenges”,IEEE Access, Vol. 3, pp. 1379-1391.
Warwick, K. (2012), Artificial Intelligence: The Basics, 1st ed., Routledge, London.
Waters, R. (2015), “Artificial intelligence: a virtual assistant for life”, available at: www.ft.com/content/
4f2f97ea-b8ec-11e4-b8e6-00144feab7de
Weber, F.D. and Schütte, R. (2019), “State-of-the-art and adoption of artificial intelligence in retailing”,
Digital Policy, Regulation and Governance, Vol. 21 No. 3, pp. 264-279.
Weir,C.S.,Douglas,G.,Carruthers,M.andJack,M.(2009),“User perceptions of security, convenience and
usability for ebanking authentication tokens”,Computers and Security, Vol. 28 Nos 1/2, pp. 47-62.
Winter, J.S. and Davidson, E. (2019), “Governance of artificial intelligence and personal health
information”,Digital Policy, Regulation and Governance, Vol. 21 No. 3, pp. 280-290.
Wirtz, B.W., Weyerer, J.C. and Geyer, C. (2019), “Artificial intelligence and the public sector –4applications
and challenges”,International Journal of Public Administration, Vol. 42 No. 7, pp. 596-615.
Zanzotto, F.M. (2019), “Viewpoint: human-in-the-loop artificial intelligence”,Journal of Artificial
Intelligence Research, Vol. 64, pp. 243-252.
Corresponding author
Umar Mir can be contacted at: mirumar.iitd@gmail.com
For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com
JSTPM
