Content uploaded by Sharmin Nahar
Author content
All content in this area was uploaded by Sharmin Nahar on Jan 18, 2024
Content may be subject to copyright.
Technological Forecasting & Social Change 201 (2024) 123203
Available online 17 January 2024
0040-1625/© 2024 The Author. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Modeling the effects of articial intelligence (AI)-based innovation on
sustainable development goals (SDGs): Applying a system dynamics
perspective in a cross-country setting
Sharmin Nahar
Lancaster University Management School (LUMS), Lancaster University, Bailrigg, Lancaster LA1 4YX, United Kingdom of Great Britain and Northern Ireland
ARTICLE INFO
Keywords:
System dynamics
Articial intelligence (AI)
Innovation
Sustainable development goals (SDGs)
Cross-country study
Institutional theory
Technology enactment framework
ABSTRACT
Global environmental outcomes, productivity, inclusion, and equality aspects are already beginning to be
impacted by articial intelligence (AI), both immediately and over time. AI is expected to have both benecial
and detrimental effects on Sustainable Development Goals (SDGs). Nevertheless, there is a lacuna in the liter-
ature regarding systematically forecasting `AI’s impact on different facets of SDGs over time in various countries.
Moreover, though existing literature has reported a correlation between AI and innovation, no prior studies have
forecast the inuence of AI-based innovation on SDG Outcomes. To ll these signicant research gaps, this study
forecasts the impact of AI-based innovation on achieving SDGs over nine years, extending from 2022 to 2030 in
22 countries (including both developed and developing countries) across ve continents via system dynamics
modeling-based simulation and grounded in Institutional Theory (Technology Enactment Framework). The
ndings exhibit varying impacts on different SDGs. This study enriches the AI, innovation, and sustainable
development literature by providing forecasts of the intricate relationship between AI, innovation, and SDGs,
thereby offering valuable insights to the reader.
1. Introduction
It is a premise of contemporary economics that the development of
new technologies is driven by a desire to improve the lives of in-
dividuals. However, human progress has resulted in environmental
deterioration, social inequity, resource depletion, disputes, and even
international wars since the beginning of the industrial revolution. For
these reasons, the United Nations (UN) and other international organi-
zations have long advocated for the concept of ‘sustainable develop-
ment,’ seeing it as a framework for achieving human progress that
strikes a good balance between economic, environmental, and social
concerns. Articial intelligence (AI) is currently being widely used with
continuous improvements in science and technology, producing a dra-
matic shift in human civilization (Johnson et al., 2022). It is, practically,
hard to block the development of increasingly powerful AI technology
due to people’s hunger for positive advances.
1
Concurrently, many are
anxious about the possibility of expanding wealth gaps and hastening
the death of our civilization (Lee et al., 2022). These scenarios are
referred to as a dilemma or a trilemma. People have either balanced or
polarized attitudes to these concerns, much like other concerns, such as
climate change, where the trend is visible and unavoidable.
As a result, academic researchers have revived their interest in AI
over the last 50 years (Di Vaio et al., 2020). AI is anticipated to impact
signicantly on global environmental outcomes, productivity, inclusion,
E-mail address: s.nahar@lancaster.ac.uk.
1
For instance, AI can optimize energy systems, improve efciency and reduce waste, enabling sustainable development. It can also help to predict natural disasters,
optimize agriculture, and manage water resources more efciently, reducing the negative impact on the environment. Additionally, AI can facilitate the development
of sustainable products and services, promoting a circular economy (Johnson et al., 2022).
Contents lists available at ScienceDirect
Technological Forecasting & Social Change
journal homepage: www.elsevier.com/locate/techfore
https://doi.org/10.1016/j.techfore.2023.123203
Received 14 December 2022; Received in revised form 29 December 2023; Accepted 29 December 2023
Technological Forecasting & Social Change 201 (2024) 123203
2
and equality. AI’s effects on sustainable development
2
have been mixed
(Lammers et al., 2022). However, no prior published study systemati-
cally predicts the extent to which AI may inuence each facet of Sus-
tainable Development Goals (SDGs) over time in different developed and
developing countries.
3
This indicates a crucial research gap, as anec-
dotal evidence indicates that AI might impact on the capacity to achieve
each SDG (Leal Filho et al., 2022). Therefore, a comparative analysis
would contribute to the extant literature by providing empirical evi-
dence on the impact of AI-based innovation on SDGs in different
developed and developing countries and identifying the factors that
inuence that impact.
Moreover, although existing literature has reported a correlation
between AI and innovation (Pietronudo et al., 2022), and between
innovation and SDGs (Abbasi et al., 2022), no prior studies have forecast
the inuence of AI-based innovation on SDG outcomes in cross-country
settings. Therefore, a critical research gap exists in this area. To address
this critical research gap, this study investigates the following research
question grounded in Institutional Theory (Technology Enactment
Framework)
4
:
What is the impact of AI-based innovation
5
on achieving SDGs over
time in developed and developing nations?
To answer this research question, this study examined how AI-based
innovation can either catalyze or impede the achievement of each of the
17 SDGs outlined in the 2030 Agenda. AI makes machines act intelli-
gently (Di Vaio et al., 2020; McCarthy et al., 2006). Thus, AI is the ca-
pacity of a system to perform tasks smartly in ever-expanding regions
and over time (Berente et al., 2021), precisely construing external data
and utilizing these lessons towards accomplishing certain aims through
an agile setup (Kaplan and Haenlein, 2020). Therefore, the research
question is answered from a system dynamics perspective
6
(Forrester,
1958; Forrester et al., 1976).
Since SDG9 represents innovation, this research also explores the
interaction between SDG9 and the rest of the SDGs separately. This is
also a unique contribution of this study as it explores the complex
interaction between SDG9 and the rest of the SDGs. It should also be
noted that this study categorizes the 17 SDGs by the three pillars of
sustainable development: social,
7
economic,
8
and environmental,
9
following Ranjbari et al. (2021).
Although each construct of this research has been studied separately
in the past, studying their interaction is novel because it presents a fresh
perspective on the impact of AI-based innovation on SDGs, which has
not been explored in the existing literature. This novel approach is
grounded in the Technology Enactment Framework (TEF), which orig-
inated in Institutional Theory (Fountain, 2001). TEF examines how
institutional, organizational, technical, and situational variables impact
on SDG attainment across countries, and this framework is used to
explore the impact of particular institutional arrangements and orga-
nizational structures on the characteristics of the technology. By
extending TEF to demonstrate the coexistence of AI and innovation in
accomplishing SDGs, the paper adds a new perspective to the existing
literature. Furthermore, the paper combines institutional theory and
system dynamics to better understand the complex interaction between
AI, innovation, and SDGs. Thus, the study’s contribution lies in
providing a theoretical framework that integrates institutional theory
and system dynamics to analyze the interaction between AI, innovation,
and SDGs. Hence, this paper makes a signicant contribution to the
extant literature on AI, Innovation, Sustainable Development, and Sys-
tem Dynamics. Apart from these theoretical contributions, this study
also provides valuable contributions to practitioners and policymakers
as a planning tool and useful guideline.
The paper continues as follows. The following section examines the
existing literature concerning the overall research theme and the vari-
ables under consideration. The section that follows establishes the
research design before presenting the data analyses. The paper con-
cludes with sections on the ndings, conclusion, and possible future
research directions.
2. Literature review and research framework
2.1. Institutional theory as the theoretical framework
Researchers are starting to understand the intricate relationship with
the social context across countries in which an AI-based innovation
environment is chosen, developed, implemented, and used (He et al.,
2022; Turr´
o et al., 2014). According to this theory, social structures and
AI have a complicated and recursive relationship that makes the results
of AI unpredictable and uncertain. These studies assert that organiza-
tional, social, and technological contexts affect AI (Munoko et al., 2020).
One of these integrative approaches is institutional theory, which ac-
knowledges the context in which articial intelligence is embedded and
explains how various institutions and factors inuence the choice,
design, implementation, and use of AI in a cross-country context (Gar-
cía-S´
anchez et al., 2020b).
Institutions have been conceptualized in various ways throughout
2
In this study, I dene sustainable development as per the 17 interconnected
SDGs from the United Nations (UN) and 169 targets internationally agreed
upon in the 2030 Agenda for Sustainable Development. These are ‘shared
blueprints for peace and prosperity for people and the planet, now and into the
future.’ These 17 SDGs are: 1) No Poverty, 2) Zero Hunger, 3) Good Health and
Well-being, 4) Quality Education, 5) Gender Equality, 6) Clean Water and
Sanitation, 7) Affordable and Clean Energy, 8) Decent Work and Economic
Growth, 9) Industry, Innovation and Infrastructure, 10) Reduced Inequality, 11)
Sustainable Cities and Communities, 12) Responsible Consumption and Pro-
duction, 13) Climate Action, 14) Life Below Water, 15) Life on Land, 16) Peace
and Justice Strong Institutions, 17) Partnerships to achieve these Goals (United
Nations, 2022a).
3
Anecdotal evidence shows that this inuence may vary across different
developed and developing countries due to differences in socio-economic con-
ditions, institutional factors, and policy environments. Developed countries
may benet more from AI-based innovation due to their better infrastructure,
more advanced technology, and higher levels of education and research.
Developing countries may struggle to adopt and adapt AI technology due to
limited resources and infrastructure, resulting in a potential widening of the
technology divide between developed and developing nations. Hence, a
comparative analysis is necessary to identify the factors that inuence the
impact of AI-based innovation on the achievement of SDGs in different devel-
oped and developing countries (Mathiyazhagan et al., 2021).
4
The Technology Enactment Framework, based on institutional theory, ex-
amines institutions, organizations, and information technologies.
5
These AI-based Innovations include the Internet of Things (IoT), Augmented
Reality, Blockchain, Virtual Reality, 5G Communication Infrastructure, Digital
Twin, Big Data, Recommender, and Information Systems.
6
System dynamics is a methodology used to analyze complex systems, it
involves the development of mathematical models that capture the dynamic
behavior of a system over time. It is particularly useful for studying systems
with feedback loops, delays, and non-linear relationships between variables
(Forrester et al., 1976). In the context of this paper, system dynamics is used to
model the complex interactions and feedback loops between AI-based innova-
tion and SDGs in a cross-country setting, taking into account differences in
social, economic, and political conditions.
7
Social categories include: 4) Quality Education, 5) Gender Equality, 10)
Reduced Inequality, 11) Sustainable Cities and Communities, 16) Peace and
Justice Strong Institutions, and 17) Partnerships to achieve these Goals (Ranj-
bari et al., 2021).
8
Economic categories include: 1) No Poverty, 2) Zero Hunger, 3) Good
Health and Well-being, 8) Decent Work and Economic Growth, and 9) Industry,
Innovation and Infrastructure (Ranjbari et al., 2021).
9
Environmental categories include: 6) Clean Water and Sanitation, 7)
Affordable and Clean Energy, 12) Responsible Consumption and Production,
13) Climate Action, 14) Life Below Water, and 15) Life on Land (Ranjbari et al.,
2021).
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
3
institutional theory’s evolution. They are regarded as guidelines for
human behavior or acceptable social conduct (Kabengele and Hahn,
2021). According to Struckell et al. (2022) and Aparicio et al. (2021),
institutions are viewed as behavioral rules based on a variety of signif-
icant foundations, including culture, social norms, mental models,
legislation, and political arrangements. These various foundations can
be summarized as cultural-cognitive, normative, and regulative pillars
that symbolize institutions in a cross-country context (Lehmann et al.,
2022).
In sustainable development discussions, institutions are crucial. Both
developed and developing nations require individuals and businesses to
protect the environment and society (Gerged and Almontaser, 2021).
This trend is supported by international treaties signed in the second
part of the 21st century. Conventions on human rights and working
conditions adopted by the UN and the International Labour Organiza-
tion (ILO) are two such examples. Ethical and sustainability issues
become more important as societal norms and values change (García-
S´
anchez et al., 2020a). Thus, institutionalism explains why people and
businesses behave socially responsibly, how different nations approach
sustainable business growth, and how this approach evolves (Barba-
S´
anchez et al., 2019; Fritz et al., 2021).
In this regard, SDGs culminate in many universally agreed-upon
rules, principles, and guidelines intended for individuals, businesses,
and normative institutional initiatives. Thus, formal and informal in-
stitutions will inuence individuals’ and businesses’ involvement with
sustainability themes like SDGs (Fritz et al., 2021; Gerged and Almon-
taser, 2021). Modgil et al. (2020) argue that Institutional Theory
benchmarks individuals’ and rms’ competitive behavior to achieve
SDGs in a cross-country context. However, the SDGs themselves may
also be seen as an initiative taken by institutions.
The Technology Enactment Framework (TEF) is a conceptual
framework, based on Institutional Theory (Fountain, 1995, 2001),
which explains how technology is implemented and used within orga-
nizations and across countries. According to the TEF, the implementa-
tion and use of technology are inuenced by institutional structures,
bureaucratic hierarchies, social norms, and information technologies’
convergence (Fountain, 2001). This framework’s underlying logic is that
‘objective technologies’ are transformed into ‘enacted technologies’
through organizational and/or individual forms
10
and institutional ar-
rangements.
11
Similarly, the choice, design, and implementation of
enacted technologies impact on organizational forms and institutional
structure in a cross-country setting (Mu et al., 2022). As a result, the
technology enactment framework recognizes the recursive nature of the
relationships between organizational, institutional, and enacted tech-
nologies. Fig. 1 illustrates the Technology Enactment Framework (TEF)
(Fountain, 2001).
In the context of this paper, I have not considered the recursive na-
ture of organizational forms, institutional arrangements, and enacted
technologies’ relationships. Rather, I have considered unidirectional
relationships among the aforementioned factors. In this paper, Objective
Technologies are AI, associated hardware and software, and all the
features (potentials) that are not always utilized. For example, multiple
hidden functionalities of AI are known and used by a limited number of
individuals. In contrast to all the features which could have been
included (Objective Technology), Enacted Technologies are those as-
pects of AI that are actually in use (they are part of the current
information system or systems). To be more precise, Enacted Technol-
ogy in this paper is Innovation, which refers to the outcome of AI use
(Objective Technologies). In terms of effectiveness, efciency, and
transparency, this innovation (Enacted Technology) results in specic
outcomes in achieving SDGs.
As mentioned earlier, Enacted technologies are affected by the
organizational forms which imply the context or social aspects of using
AI (e.g., whether AI implementation is a bureaucratic/straightforward
process, the social capital
12
in using AI). These Organizational Forms
(the context of AI utilization) are catalyzed or hindered by institutional
arrangements in a country symbolized by laws, rules, and other cultural,
cognitive, or socio-structural controls in this study.
A literature review on each of the aforementioned constructs and
their interrelationship is conducted below:
2.1.1. Articial intelligence (AI)
Even though the study of Articial Intelligence (AI) has been going
on for the last six decades, it is still challenging to come up with a
consensus on the denition of this ‘set of technologies.’ In fact, the term
‘AI’ refers to diverse technologies, for instance, robotics, deep learning,
machine learning, speech recognition, image recognition, computer
vision, natural language processing, and analytics that mimic human
behavior, rather than a single technology or even a class of technologies
(Huang and Rust, 2018).
Huang and Rust (2018) claim that AI is a machine with intelligence.
Truong and Papagiannidis (2022) dene AI as computer intelligence
that mirrors cognitive abilities, including problem-solving. AI was
dened by Longoni et al. (2019) as any computer that utilizes any al-
gorithm or numerical method to execute human-like cognitive,
perceptual, and conversational activities.
AI, in contrast to older manufacturing or information technologies, is
able to learn from data, process data for human use, and update out-
comes without the need for programming or human involvement
(Huang and Rust, 2018). For instance, AI researchers have created
potent algorithms that can assist social scientists in overcoming the
difculties associated with analyzing ‘big’ behavioral data, such as SDGs
(Leal Filho et al., 2022).
The above literature review shows that Articial Intelligence (AI) is a
rapidly evolving technology that has revolutionized how we live and
work. AI is the development of computer systems that can perform tasks
that would typically require human intelligence. AI has been at the
forefront of many innovative breakthroughs in recent years and has been
found to impact signicantly on innovation in various industries
(Truong and Papagiannidis, 2022).
Fig. 1. Technology Enactment Framework (Fountain, 2001).
10
Structural dimensions such as centralization, formalization, communication
channels, and organizational bureaucratic characteristics are examples of
organizational forms (Gil-Garcia, 2005).
11
Institutional arrangements shape users’ perceptions and behaviors towards
technology, including cognitive institutions, culture, socio-structure, and legal
norms. Cognitive institutions are mental routines and cognitive models that
affect behavior and decision-making. Cultural institutions are characterized as
shared symbols, such as stories or meanings (Fountain, 2001).
12
Social capital refers to the interconnected relationships among individuals
within a particular group or society, which facilitate its effective functioning
(Akizu-Gardoki et al., 2018).
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
4
2.1.2. AI and innovation
Innovation is crucial for the growth of businesses in today’s rapidly
changing world because innovation creates valuable products, services,
and processes. The innovation process typically consists of four stages:
1) concept discovery and creation; 2) ltering; 3) testing; and 4)
renement and commercialization of the winning idea (Truong and
Papagiannidis, 2022). While humans have exceptional problem-solving
skills, AI has advanced to the point where it can do complex tasks, and it
has the potential to streamline innovation by processing vast volumes of
data and information that innovators must navigate (Cockburn et al.,
2019). By automating tasks such as categorization, grouping, and pre-
diction, AI can help innovators focus on more mentally taxing but less
judgment-heavy support activities. This can reduce the time and money
spent on processing data and enable innovators to spend more time on
creative problem-solving (Cockburn et al., 2019).
AI can also be used to improve the innovation process by providing
insights and recommendations based on data analysis. AI, for instance,
can be used to analyze customer feedback and determine areas of a
product or service that might benet from being improved. According to
Haefner et al. (2021), this can assist innovators in rening their ideas
and developing products and services that more effectively meet the
needs of customers.
AI can also enhance collaboration among innovators. By providing
real-time insights and data analysis, AI can help team members stay
informed and make more informed decisions. This can lead to better
communication and collaboration among team members, which can
help to speed up the innovation process (Cockburn et al., 2019).
Despite the above positive impact of AI on innovation, it also poses
some challenges for innovation. For instance, while AI has advanced to
the point where it can do complex tasks, it is still largely utilized for
niche applications that need careful human oversight. Humans are still
needed as the source of fresh ideas and innovations, as creativity and
problem-solving skills are not easily replicated by AI (Amabile, 2020).
By leveraging AI’s strengths and addressing the aforementioned con-
cerns, innovators can use AI to drive innovation and create valuable
products, services, and processes.
2.1.3. Sustainable development goals (SDGs)
Fifty years of discussion and agreement on sustainable development
led to SDGs. To help move the world onto a more sustainable and
resilient path, global leaders at the UN Sustainable Development Sum-
mit in 2015 endorsed the 2030 Agenda for Sustainable Development,
which they call ‘a plan of action for people, planet, and prosperity.’ This
universal, integrated, and transformative agenda is based on 17 SDGs,
169 targets, and 232 indicators (United Nations, 2022a). Fig. 2 depicts
all 17 SDGs organized into three pillars: environmental, social, and
economic.
The new SDGs framework, which is more ambitious than the Mil-
lennium Development Goals (MDGs), includes bolder goals, such as the
end of poverty (SDG1), zero hunger (SDG2), good health and well-being
for all (SDG3), quality education (SDG4), gender equality (SDG5), clean
water and sanitation (SDG6), access to affordable, sustainable, and clean
energy (SDG7), decent work and economic growth (SDG8), industry,
innovation and infrastructure (SDG9), reduced inequalities (SDG10),
sustainable cities and communities (SDG11), responsible consumption
and production (SDG12), actions to combat climate change (SDG13),
protecting and promoting sustainability for life below water (SDG14),
protecting life on land (SDG15), peace, justice and strong institutions
(SDG16), and revitalizing the global partnership for sustainable devel-
opment (SDG17) (United Nations, 2022a). The SDGs balance economic,
social, and environmental sustainability. They are unique in requiring
all nations and their citizens to reduce inequalities. National govern-
ments, transnational companies, and civil society can end poverty,
promote economic growth, and address social and environmental needs
through the Goals. The 2030 Agenda also encourages private sector and
partnership involvement to help governments use all the tools needed to
implement and deliver change (United Nations, 2022a).
2.1.4. AI’s role in achieving SDGs through innovation
In accordance with the SDGs, AI-based innovation is currently being
used to address the world’s most signicant issues. These powerful
forces are bringing about change in both the business world and society
by forming public-private partnerships that bring together a wide range
of specialized knowledge. AI is already being applied to all 17 SDGs
(GSMA, 2022). Specically, AI promotes SDG9, which embraces three
essential attributes of sustainable development: infrastructure, indus-
trialization, and innovation.
2.1.4.1. AI in achieving the economic pillar of SDGs through innovation.
According to the technology enactment framework of institutional the-
ory, SDG9 (AI-based innovation) contributes to the success of the eco-
nomic pillar of the SDGs (SDG1–3 and SDG8), which focuses on creating
sustainable, inclusive, and equitable economic growth and development
(Mathiyazhagan et al., 2021). For instance, AI-based innovation (such as
digital nancial services) stimulates economic growth, creates employ-
ment opportunities (SDG8), and reduces poverty (SDG1) in cross-
country settings, as indicated by several prior studies. First, AI (objec-
tive information technology) will boost global GDP by 14 % to $15.7
trillion in 2030, more than China and India combined (World Economic
Forum, 2023). AI could generate between $3.5 and $5.8 trillion in
annual value across various industries. This increased value can be
attributed to AI’s ability to automate repetitive tasks, optimize pro-
cesses, and improve decision-making (Truong and Papagiannidis, 2022).
Secondly, AI-based innovation (enacted technology) can lead to job
creation and upskilling. Contrary to the popular belief that AI will lead
to massive job displacement, several studies have suggested that AI-
based innovation can create new jobs and increase demand for skilled
labor. For instance, PricewaterhouseCoopers (PwC) estimates that AI
could create approximately 7.2 million new UK jobs by 2037 (Kollewe,
2023). Moreover, AI can also facilitate upskilling and reskilling of the
workforce by providing personalized training and learning opportu-
nities in different countries (Kollewe, 2023). Thirdly, AI-based innova-
tion can promote inclusive economic growth by reducing inequalities
and improving access to services. For instance, AI can improve access to
nancial services by automating credit scoring and underwriting pro-
cesses, thus reducing bias and improving nancial inclusion. Addition-
ally, SDG9 (such as innovative food storage and preservation strategies,
sustainable crop production, and sustainable livestock farming) facili-
tates the achievement of SDG2 (Zero Hunger by 2030). In addition,
SDG9 (e.g., innovative health technologies) is indispensable for
improving health and lowering infant, maternal, and adult mortality
rates (SDG3) (United Nations, 2022c). Consequently, the following is
hypothesized:
H1. AI-based innovation (SDG9) positively inuences the SDGs’ eco-
nomic pillar
13
(i.e., SDG1-3 and SDG8).
2.1.4.2. AI in achieving the social pillar of SDGs through innovation. Ac-
cording to the technology enactment framework of institutional theory,
SDG9 (AI-based Innovation) also contributes to the success of the SDGs’
social pillar (SDG4-5, SDG10-11, and SDG16-17) in cross-country set-
tings (Mathiyazhagan et al., 2021). For instance, AI-based innovation
can positively inuence SDG4 by making education more accessible,
personalized, and inclusive. With online learning platforms, adaptive
learning systems, and intelligent tutoring systems, AI-powered tech-
nologies (objective technology) can cater to the diverse needs of learners
and provide personalized assessments, feedback, and real-time learning.
This innovation also enables inclusive and equitable education,
13
The four associated sub-hypotheses are provided in Table 18 in the
Appendix.
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
5
supporting learners with disabilities and providing education access to
remote and underserved areas (Lammers et al., 2022). AI-based inno-
vation (enacted technology) can also support professional development
for educators, provide personalized training and mentoring, and analyze
data to identify learning gaps and opportunities for improvement in
different countries (United Nations, 2022d). In conclusion, AI-based
innovation can enhance the quality of education and support
evidence-based decision-making, contributing positively to SDG4.
AI-based innovation can also positively inuence SDG5, which aims
to achieve gender equality and empower all women and girls. Through
safety apps, online campaigns, sensor-laden smart dresses, and AI-
powered diversity and inclusion tracking platforms, AI-based innova-
tion can enhance women’s safety, promote their rights, and advance
gender equality (United Nations, 2022c). AI-powered technologies can
also provide insights into gender-based biases and discrimination,
helping policymakers develop evidence-based policies to tackle these
issues. By reducing the gender gap in access to technology, AI-based
innovation can facilitate women’s participation in the digital economy
and improve their economic opportunities (de Sousa Jabbour et al.,
2020). Overall, AI-based innovation can promote gender equality and
women’s empowerment, contributing positively to SDG5.
AI-based innovation can also positively inuence SDG10, which aims
to reduce inequalities within and among countries. AI-based innovation
can help to narrow the digital divide and promote social and economic
inclusion by facilitating access to technologies and knowledge for
disadvantaged segments of society. AI-powered technologies can also
provide personalized services and solutions, improving access to
healthcare, education, and other essential services for marginalized
communities (Cosenz et al., 2020). By analyzing data, AI-based inno-
vation can also help to identify and address inequalities in different
domains, such as income, education, and health. In addition, AI-based
innovation reduces inequality within and between nations, commu-
nities, and populations by facilitating access to technologies and
knowledge for disadvantaged segments of society (SDG10) (Fritz et al.,
2021). Overall, AI-based innovation can contribute to reducing in-
equalities and promoting social and economic inclusion, supporting the
achievement of SDG10.
AI-based innovation can also positively inuence SDG11, which aims
to make cities and human settlements inclusive, safe, resilient, and
sustainable. Through 3D-scanning technology, AI-based decision-mak-
ing for urban planning, green public transport, and AI-powered air
quality platforms, AI-based innovation can contribute to building smart,
sustainable, and livable cities (de Sousa Jabbour et al., 2020). By opti-
mizing resource allocation, improving infrastructure, and enhancing
urban mobility, AI-based technologies can improve the quality of life for
urban residents, while reducing the environmental impact. Further-
more, AI-based innovation can support disaster response and manage-
ment, ensuring the resilience of urban areas (Lammers et al., 2022).
Overall, AI-based innovation can contribute to creating sustainable and
inclusive cities, supporting the achievement of SDG11.
AI-based innovation can also positively inuence SDG16, which aims
to promote peaceful and inclusive societies for sustainable development,
provide access to justice for all, and build effective, accountable, and
inclusive institutions at all levels. Through AI-powered data analysis and
decision-making, AI-based innovation can help to promote trans-
parency, accountability, and good governance, supporting the estab-
lishment of effective and inclusive institutions (García-S´
anchez et al.,
2020a). By analyzing data on crime and security, AI-based technologies
can also support crime prevention and improve public safety, promoting
peaceful societies. Furthermore, AI-based innovation can also support
conict resolution and reconciliation efforts by analyzing social media
data and other sources of information (Gerged and Almontaser, 2021).
Overall, AI-based innovation can promote peace, justice, and inclusive
institutions, supporting the achievement of SDG16.
AI-based innovation can also positively inuence SDG17, which aims
to strengthen the means of implementation and revitalize the global
partnership for sustainable development. Through virtual meeting
platforms, AI-based decision-making for partnering, and other AI-
powered technologies, AI-based innovation can facilitate public-
private partnerships that mobilize all available resources for sustain-
able development by bringing together the Government, private sector,
and civil society (Abbasi et al., 2022). By optimizing resource allocation
and enhancing collaboration between different sectors, AI-based tech-
nologies can also improve the effectiveness and efciency of develop-
mental cooperation. Furthermore, AI-based innovation can provide
insights into the impact of different policies and interventions, enabling
evidence-based decision-making for sustainable development (García-
S´
anchez et al., 2020a). Overall, AI-based innovation can contribute to
revitalizing the global partnership for sustainable development, sup-
porting the achievement of SDG17. Consequently, the following is
hypothesized:
H2. AI-based innovation (SDG9) positively inuences the SDGs’ social
pillar
14
(i.e., SDG4-5, SDG10-11, and SDG16-17).
2.1.4.3. AI in achieving the environmental pillar of SDGs through
innovation. According to the technology enactment framework of insti-
tutional theory, the environmental pillar of the SDGs (SDG6-7 and
Fig. 2. 17 SDGs grouped into three pillars (triple bottom line): environmental, social, and economic (adapted from United Nations, 2022a).
14
The six associated sub-hypotheses are provided in Table 18 in the Appendix.
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
6
SDG12-15) also benets from SDG9 (AI-based innovation) in cross-
country settings (Mathiyazhagan et al., 2021). For instance, AI-based
innovation can positively inuence SDG6, which aims to ensure the
availability and sustainable management of water and sanitation for all.
Through the analysis of data on water quality and usage, AI-based
technologies (e.g., the water sector’s digital transformation) can sup-
port the optimization of water supply and sanitation systems, improving
access to clean water and sanitation services. Furthermore, AI-based
innovations can also support the management of water resources and
monitoring of water-related risks, such as oods and droughts (de Sousa
Jabbour et al., 2020). Overall, AI-based innovation (enacted technol-
ogy) can contribute to improving water management and increasing
access to clean water and sanitation worldwide, supporting the
achievement of SDG6.
AI-based innovation can also positively inuence SDG7, which aims
to ensure access to affordable, reliable, sustainable, and modern energy
for all. Through analyzing energy data, AI-based technologies (objective
technology) can support the optimization of energy generation, distri-
bution, and consumption, reducing waste and increasing efciency.
Furthermore, AI-based innovations can also support the integration of
renewable energy sources such as solar, wind, geothermal, and off-grid
energy systems into existing energy systems, facilitating the transition to
a more sustainable energy mix (Ranjbari et al., 2021). Overall, AI-based
innovation can contribute to increasing access to affordable and sus-
tainable energy, supporting the achievement of SDG7 worldwide.
AI-based innovation can also positively inuence SDG12 (Respon-
sible Consumption and Production) by improving resource efciency,
reducing waste and pollution, and promoting sustainable practices. AI-
based innovation (e.g., food system innovations such as automation
and robotics in agriculture and personalized nutrition) can also enable
more effective supply chain management, reduce the environmental
impact, and promote circular economy models, thereby contributing to
SDG9 and promoting sustainable development (Fritz et al., 2021).
AI-based innovation (e.g., innovations to combat climate change
such as innovative low-carbon products and services, AI-based green
data centers, sophisticated climate modeling, and green power feeding
systems) can also positively inuence SDG13 (Climate Action) by
facilitating the development and deployment of clean energy solutions,
enhancing climate modeling and prediction, improving climate resil-
ience, and supporting carbon capture and storage technologies. AI can
also help to identify new opportunities for emissions reductions and
enable more effective climate monitoring and mitigation strategies,
contributing to achieving SDG9 and promoting sustainable development
(Richter et al., 2022).
AI-based innovation can also positively inuence SDG14 (Life Below
Water) by enhancing ocean monitoring and modeling, supporting sus-
tainable sheries management, and improving marine biodiversity
conservation. AI can also enable real-time data collection and analysis,
identify threats to marine ecosystems, and support effective response
measures, contributing to achieving SDG9 and promoting sustainable
development in different countries (Ranjbari et al., 2021).
AI-based innovation can also positively inuence SDG15 (Life on
Land) by improving land-use mapping, enhancing wildlife conservation,
and supporting sustainable forest management. AI can also enable more
accurate biodiversity monitoring, identify and mitigate the impacts of
human activities, and support ecosystem restoration efforts, contrib-
uting to the achievement of SDG9 and promoting sustainable develop-
ment in cross-country settings (de Sousa Jabbour et al., 2020).
Therefore, the following is hypothesized:
H3. AI-based innovation (SDG9) positively inuences the SDGs’
environmental pillar
15
(i.e., SDG6-7 and SDG12-15).
2.1.5. Impact of cross-country institutional arrangements on the
relationship between AI-based innovation and SDGs’ achievement
When governments use information and communications technology
(ICT), the results vary (Richter et al., 2022). For instance, many coun-
tries have successfully collected income taxes online. However, new ICT
systems sometimes fail to improve processes, data, or knowledge-
sharing. Researchers and practitioners want to understand these com-
plexities (Richter et al., 2022). The technology enactment framework
(TEF) may help to understand these complexities (Fountain, 2001).
‘Building the Virtual State’ by Jane Fountain examines ICT’s impact on
government institutions. Fountain uses bureaucracy, neo-
institutionalism, networks, and governance research to present the
technology enactment framework. It offers a more complete and
convincing explanation than her partially developed theories. This al-
lows an in-depth study of how organizational structure and institutional
arrangements affect technology use for theory development. Fountain’s
conceptual separation of ICT components from actors’ interpretations
and practical applications is crucial. According to the TEF, technology
serves three purposes: it manages the organization, becomes part of its
infrastructure, and drives organizational change. Fountain refers to ICT-
heavy states as ‘virtual states,’ a metaphorical term used to underscore
their impacts on state structure, citizens, and corporations (Fountain,
2001).
Apart from shedding light on formal institutions such as Govern-
ment, TEF acknowledges the role of informal institutions such as un-
written rules, social norms, and shared beliefs in shaping how
individuals, organizations, and societies adopt and use AI-based in-
novations (Aparicio et al., 2021). Informal institutions can impact on
how people perceive and interpret AI, the level of trust they have in it,
and the rules and regulations that govern its use. By incorporating
informal institutions in TEF, researchers can better understand how AI-
based innovations can be adopted, scaled up, and sustained in different
social contexts. Informal institutions can also impact on how individuals
and organizations approach sustainability and ethical issues (Aparicio
et al., 2021). For example, shared beliefs and norms can shape people’s
perceptions and values vis-`
a-vis SDGs, inuencing the extent to which
they adopt sustainable practices and technologies. Understanding the
role of informal institutions can help individuals and institutions to
adopt AI-based innovations in a way that aligns with their values and
norms and promotes sustainable and ethical practices (Styrin et al.,
2022).
The countries (mostly developed countries) that adopted the afore-
mentioned institutional arrangements have made signicant strides
since 2015 in achieving SDGs through AI-based innovation. These
developed countries have included SDGs in national plans, strategies,
planning processes, parliaments, and multi-stakeholder institutions.
These countries have also created high-level coordination mechanisms,
dedicated strategies and roadmaps, data platforms, and collaborative
reporting to engage stakeholders. Moreover, though many developed
countries have failed to meet SDGs related to income inequality, edu-
cation, climate change mitigation, and gender equality due to the lack of
relevant innovations, they have, overall, performed well in economic
development (World Economic Forum, 2017). Furthermore, as gaps or
weaknesses emerge and circumstances change, these countries have
implemented or adjusted key institutional systems concerning
innovation-inuenced SDG implementation. Conversely, developing
countries, including the poorest, rank low in institutional arrangements,
AI-based innovation, and SDG achievement (Dhahri et al., 2021).
Consequently, the following is hypothesized:
H4. AI-based innovation helps developed countries to achieve better
results in SDGs
16
due to better institutional arrangements than
15
The six associated sub-hypotheses are provided in Table 18 in the Appendix.
16
The sixteen associated sub-hypotheses are provided in Table 18 in the
Appendix.
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
7
developing countries.
As per the above literature review, the theoretical and research
models of this study are provided below in Figs. 3 and 4, respectively:
3. Research methodology
3.1. System dynamics (SD)
This research uses system dynamics (SD), established by Forrester in
the 1950s and ‘60s. It simulates complex social systems to test new
management and change methods. SD holds that ‘effects may inuence
causes,’ requiring business analysts to consider the entire system of
interrelated relationships. SD also recognizes complex interconnected
feedback loops and rejects a linear cause-and-effect relationship (Cosenz
et al., 2020).
SD simulates differently by modeling a system’s behavior and
element relationships by mapping its structure. It helps to understand
and exploit complex management systems’ feedback relationships. The
model also provides a business planning or operational methodology for
decision-making. Furthermore, the model can test various scenarios and
predict future outcomes using historical data (Gao and Zhang, 2022).
3.2. System dynamics and sustainable development: A 50-year path
System dynamics is a natural t for sustainable Development anal-
ysis because it enables researchers to model and simulate the behavior of
complex, interconnected systems over time. System dynamics is
particularly relevant for sustainable development, which involves inte-
grating economic, social, and environmental factors to achieve long-
term, equitable outcomes (Pedercini et al., 2020). System dynamics fa-
cilitates well-analyzed policy development by identifying unintended
consequences and trade-offs. It also encourages interdisciplinary
collaboration, leading to a more comprehensive understanding of sus-
tainable development (Randers, 2000). As shown in Fig. 5, the two ‘SDs’
have coevolved over the last 50 years due to a growth-generating
feedback loop in which better analytical tools discover answers and
new problems, which in turn encourages further methodological ad-
vances (Pedercini et al., 2020).
3.3. Research setting
The impact of AI-based innovation on SDGs is simulated over a 9-
year period, from 2022 to 2030, for 22 nations (including both devel-
oped and developing nations
17
) across ve continents (i.e., Asia, Africa,
Europe, North America, and Oceania). According to the United Nations
Development Programme’s Human Development Index (HDI),
18
these
22 nations are either developed or developing (UNDP, 2022). The
countries that rate as having ‘very high human development’ in this
index have been considered as ‘developed,’ and the rest of the countries
as ‘developing.’
3.4. Problem identication
The procedure for system dynamics modeling, as outlined by
Kazancoglu et al. (2021), starts with a comprehensive understanding of
the system under examination, followed by the demarcation of its
boundaries
19
and components. Within the framework of this study, a
thorough consideration of all essential variables that inuence the sys-
tem is indispensable for accurately simulating the impacts of AI-based
innovation on SDGs. Once the system parameters (variables or con-
structs) had been outlined, their interconnections were demonstrated
through causal loop diagrams, as discussed by Kazancoglu et al. (2021).
These causal loop diagrams represent the system’s internal interactions
via positive or negative feedback loops. Analyzing these loops offers
insights into the system’s overall conduct, as evidenced in studies by
Yuan and Wang (2014) and Ricciardi et al. (2020). Afterwards, the
causal loop diagram underwent a transformation into a stock-and-ow
diagram, with the help of Vensim software, to estimate parameter
values. The subsequent sections outline the execution steps.
3.5. Sample and data
To simulate the impacts of AI-based innovation on SDGs, I employed
a comprehensive approach, drawing upon data from three distinct yet
complementary data sets. The objective was to capture all the critical
factors inuencing the system, thereby offering a multifaceted
perspective on the relationship between AI, innovation, and
sustainability.
The selection of countries for the study was vital to ensure a broad
representation across the development spectrum. I included a diverse
mix of 22 countries from both developed and developing nations. This
selection enabled me to conduct comparative analyses and to account
for the unique intricacies of AI-based innovation and SDG achievements
across the globe. This selection not only offers a globally relevant
perspective but also captures the context-specic nuances of AI-driven
sustainable development, given the 22 countries’ unique socio-
economic characteristics and stages of development (Burchardt and
Ickler, 2021; Qureshi, 2022).
The developed countries examined in this study include Australia,
Canada, France, Germany, Japan, New Zealand, Russia, Saudi Arabia,
Turkey, the UAE, the UK, and the US. These nations are often seen as
leaders in technological adoption and provide a rich context for un-
derstanding how developed economies utilize AI-based in pursuit of
their SDGs (Oxford Insights, 2022; United Nations, 2022a; WIPO, 2022).
In contrast, the study also encompasses developing nations, such as
Bangladesh, China, Egypt, India, Iran, Kenya, Nigeria, Pakistan, Uzbe-
kistan, and Vietnam. While these countries face different challenges and
are at varied stages of technological adoption, they have embarked on
ambitious plans to incorporate AI when addressing their unique devel-
opmental goals (Oxford Insights, 2022; United Nations, 2022c; WIPO,
2022).
The Oxford Insights AI readiness survey, an integral part of my data
gathering, provides a comprehensive assessment of a country’s pre-
paredness to adopt and optimally exploit articial intelligence (AI) in
the domain of public service delivery. It employs a diverse set of in-
dicators, such as existing infrastructure, talent base, regulatory frame-
works, and the level of investment in AI-related research and
development. Existing infrastructure evaluates digital resources,
internet connectivity, and data centers (Oxford Insights, 2022). The
17
A developed nation has a high quality of life, a strong economy (e.g., high
GDP, GNP, and per capita income), and an advanced technological infrastruc-
ture relative to less industrialized nations (IGI Global, 2022). On the other
hand, a country with a lower HDI and a smaller industrial base is considered a
developing nation (Bhimani et al., 2022).
18
While GDP per capita is a common indicator for categorizing countries as
developed or developing, I opted for the HDI index for several reasons. The HDI
index takes into account not only economic indicators but also social and
human development factors, such as health and education, which provides a
more comprehensive view of a country’s development status (Bhimani et al.,
2022). In this paper, I focus on all the SDGs (not only those related to the
economic pillar). Hence, the UN HDI index is the index best-suited for this
paper. Additionally, the HDI index is widely recognized and used by the United
Nations Development Programme (UNDP) as a measure of development,
making it a relevant choice for this study (UNDP, 2022).
19
In system dynamics modeling, ‘boundaries’ refer to the dened scope or
limits of the system being analyzed. Determining the boundaries of a system is a
crucial step in the modeling process, as it decides what is included within the
system (i.e., variables, relationships) (Black, 2013).
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
8
talent base gauges the quantity and quality of AI skills available,
including AI researchers, data scientists, and AI educators, in addition to
the general population’s digital literacy. Regulatory frameworks
encompass data privacy laws and AI-specic regulations, assessing the
adaptability of a country’s legal and regulatory systems in relation to the
emergence of AI. The level of investment in AI-related research and
development indicates the nancial commitment to AI advancement
through public and private sector investment (Oxford Insights, 2022).
Therefore, the survey offers a comprehensive view of each country’s AI
readiness, providing critical insights into AI adoption for policymakers,
business leaders, and other stakeholders. Beyond data collection, the
survey emphasizes the need for global efforts to bridge disparities in AI
readiness, contributing to the worldwide promotion of equitable AI
adoption (Oxford Insights, 2022).
The SDG cross-country survey by the UN, which offers a global
overview of progress on implementing the 2030 Agenda for Sustainable
Development, has been used to gather SDG-related data. The survey
gathers data from countries around the world on a range of indicators
related to SDGs, including poverty reduction, gender equality, climate
action, and economic growth (United Nations, 2022a). The objective is
twofold: to render a broad-based image of individual nations’ strides
towards these global goals, and to understand the complex interplay of
socio-economic, political, environmental, and cultural factors inu-
encing such progress. Serving policymakers, researchers, and a broad
spectrum of stakeholders, this survey describes a comprehensive land-
scape of the rate and quality of SDG advances, offering valuable insights
into what drives success in different geopolitical environments (United
Nations, 2022c). More than a simple inventory, the SDG cross-country
Fig. 3. Theoretical model of the study based on the Technology Enactment Framework (Fountain, 2001).
Fig. 4. Research model of the study.
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
9
survey is a vibrant knowledge platform, merging global accountability
for the 2030 Agenda commitments with an evolving repository of
effective strategies for sustainable development, all while underscoring
the indispensable role of international collaboration in tackling these
shared challenges (United Nations, 2022a).
The World Intellectual Property Organization (WIPO)’s Global
Innovation Index, an annual ranking of nations based on their capacity
for and success in innovation, has been used to collect innovation-
related data. This index, comprehensive in its approach, scrutinizes
the innovation environment of each country, factoring in various di-
mensions, such as expenditure on research and development, the
strength of intellectual property protection laws, and the quality of
human capital, which includes education and training (WIPO, 2022).
What sets the Global Innovation Index apart is its dual-purpose
approach: it not only highlights exemplary practices that promote
innovation and creativity, but also pinpoints areas that require
improvement. As such, it serves as a roadmap for countries to enhance
their innovation ecosystem (WIPO, 2022). The ultimate aim is to nurture
a worldwide culture of innovation and creativity, underpinning the
belief that innovation is a pivotal driver of sustainable economic growth
and development. By offering an objective analysis of each country’s
innovation strengths and weaknesses, the index plays a fundamental
role in shaping national and international policies aimed at promoting
the robustness of the global innovation landscape (WIPO, 2022).
3.6. Measures of variables (model parameters)
This multifaceted study gathers and examines data from several
variables spanning three key categories: SDGs, AI Readiness, and
Innovation.
First, the SDGs, which are universally adopted by all United Nations
Member States, form a set of objectives that strive to address a wide
spectrum of societal challenges. Their purpose is to foster a better, more
sustainable future for everyone by the year 2030. The SDGs encompass
17 interconnected goals (United Nations, 2022a). For instance, SDG1
aims to eradicate extreme poverty, dened as individuals living on less
than $1.25 per day and uses indicators such as the poverty headcount
ratio and the poverty rate following taxes and transfers to measure
progress. SDG2, titled Zero Hunger, strives to ensure that all people have
access to safe, nutritious, and sufcient food all year round, with prog-
ress assessed by indicators like the prevalence of undernourishment and
the sustainable management of nitrogen, among others (United Nations,
2022b). The remaining 15 SDGs are dedicated to other critical global
objectives. Good Health and Well-being (SDG3) ensures healthy lives
and promotes well-being for everyone at all ages. Quality Education
(SDG4) aims for an inclusive and equitable quality education and pro-
motes lifelong learning opportunities for all. Gender Equality (SDG5)
aims to achieve gender equality and empower all women and girls.
Clean Water and Sanitation (SDG6) works towards ensuring the avail-
ability and sustainable management of water and sanitation for
everyone (United Nations, 2022a). Affordable and Clean Energy (SDG7)
is focused on ensuring access to affordable, reliable, sustainable, and
modern energy for all. Decent Work and Economic Growth (SDG8)
promotes sustained, inclusive, and sustainable economic growth, full
and productive employment, and decent work for everyone. Industry,
Innovation, and Infrastructure (SDG9) aims to build resilient infra-
structure, promote inclusive and sustainable industrialization, and fos-
ter innovation. Reduced Inequality (SDG10), Sustainable Cities and
Communities (SDG11), Responsible Consumption and Production
(SDG12), Climate Action (SDG13), Life Below Water (SDG14), Life on
Land (SDG15), Peace and Justice, Strong Institutions (SDG16), and
Partnerships to achieve the Goals (SDG17) also follow suit in this regard
(United Nations, 2022c).
The second category, AI Readiness, evaluates a country’s prepared-
ness to implement AI in public services (Oxford Insights, 2022). This
evaluation takes into account an array of indicators, including the
Government’s vision for AI, the presence of regulatory and ethical AI
frameworks, the level of digital capacity within the Government, and the
size and health of the technology sector. Other considerations include
the skills available in the country, the state of technological infrastruc-
ture, and the availability of data for training AI models (Oxford Insights,
2022).
The nal category, Innovation, utilizes the Global Innovation Index
to evaluate nations based on their capacity for and success in fostering
Innovation. This category is subdivided into seven critical domains: in-
stitutions, human capital and research, infrastructure, market sophisti-
cation, business sophistication, knowledge and technology outputs, and
creative outputs (WIPO, 2022). Each domain consists of various
measurable factors, including the political environment, education
level, Research & Development investment, credit accessibility, the
availability of knowledge workers, knowledge creation and diffusion,
and intangible assets. Each of these areas provides a piece of the overall
puzzle of a nation’s capacity for innovation (WIPO, 2022).
3.7. Parameter scores
For the base year of 2021, initial values of the aforementioned model
variables were gathered. Consideration was given to the change rates of
various variables based on the current state and the 2030 vision. The
model variables are fully described in Table 1 in the Appendix.
3.8. Causal loop diagram of the model
Having dened the system’s boundaries and parameters, I developed
a causal loop diagram, as per Cernev and Fenner (2020), to investigate
the interactions between associated variables and develop a system
dynamics model. This diagram, grounded in the variables from Section
3.6 and presented in Fig. 6, offers a qualitative view of the intricate
relationship between AI-based innovation and the SDGs. In the diagram,
reinforcing effects are represented by black arrows and denoted by ‘R,’
Fig. 5. A timeline of sustainable development and system dynamics (adapted from Pedercini et al., 2020).
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
10
while balancing loops are symbolized by red arrows and labeled as ‘B.’
Notably, the diagram represents a generic causal loop diagram for 16
SDGs.
3.9. Model and data analysis procedure
To create quantitative models (stock-ow diagrams), all the major
factors that impact on the behavior of the SDGs through AI-based
innovation are identied. Stock-ow diagrams are rooted in system
dynamics, a methodology developed in the mid-20th century by Jay W.
Forrester of MIT (Forrester et al., 1976). These diagrams allow us to
visualize and understand the structure and behavior of complex systems,
such as the interplay between AI-based innovation and SDGs.
Stock, ow, converter, and connector are the four main elements of
stock-ow diagrams, representing the variables under consideration and
their equations (Cosenz et al., 2020). First, ‘Stocks’ in a system dynamic
model represent accumulations over time, such as population score-
cards, water in a bathtub, books in a library, or money in a bank. In this
research, the stock signies SDG scores (goals 1–8 and 10–17) at any
given moment. They can be understood as the ‘memory’ of the system,
because they accumulate the net effect of inows and outows over
time, showing the present state of each SDG (Gao and Zhang, 2022).
However, these stocks are not static. They change over time due to
‘ows.’ A ow is an activity that changes the stock, either by adding to it
(inows) or subtracting from it (outows). For example, births represent
the population’s lling, and deaths its draining. When applied to the
SDGs, ow could signify the rate of progress or setbacks in achieving
these goals (increase and decrease rates of SDGs). For instance, an inow
might be new policies that accelerate progress towards an SDG, while an
outow could represent factors hindering progress, like a conict or an
economic downturn (Pedercini et al., 2020). Then, the ‘converter’ (also
known as ‘auxiliary variables’) in the model symbolizes the external
inputs or variables that affect these ows. In system dynamics modeling,
auxiliary variables are split into control and main variables. ‘Control
variables,’ often adjustable parameters, inuence system behavior,
while the ‘main variables’ represent the model outputs of primary in-
terest (Richardson, 1991). In this context, AI-based innovation repre-
sents the main variable (converter), which signies the transformative
potential of AI technology to change the rate of progress of SDGs. For
example, AI’s role in optimizing resource use could accelerate the
achievement of SDG12 (Responsible Consumption and Production)
(Ricciardi et al., 2020). Conversely, factors other than AI-based inno-
vation that either boost or hinder the growth rate of each SDG can be
classied as control variables (converter). Lastly, the ‘connector’ rep-
resents the interactions and relationships between these elements. It
draws out the interdependencies between the stocks, ows, and con-
verters, creating a comprehensive model of how each part of the system
inuences the others (Yuan and Wang, 2014).
Figs. 7–22 depict stock-ow diagrams (conceptual frameworks) for
each SDG apart from SDG9 (Industry, Innovation, and Infrastructure)
using VENSIM software. VENSIM software is one of the best tools for
system dynamic modeling due to its user-friendly interface, powerful
simulation engine, versatility, robustness, and support and resources. Its
interface is easy to use, and its equations are straightforward, making it
suitable for building models quickly, even for complex systems. The
simulation engine is fast and efcient, even for large models with many
variables and feedback loops. VENSIM’s versatility allows it to be used
for various applications. Its robustness allows it to handle real-world
problems. Moreover, the helpful user community and online resources
provide a wealth of information for users (Del Vecchio et al., 2019).
Overall, VENSIM is an excellent tool for system dynamic modeling.
In the SDG1 stock-ow diagram (see Fig. 7, below), the ‘stock’ rep-
resents the current state of poverty. This is impacted by two main
‘ows’: the ‘inow,’ or a rise in the SDG1 score, reecting anti-poverty
actions, and the ‘outow,’ or a decrease in the score, indicating factors
that exacerbate poverty. The inow is positively inuenced by both the
‘main auxiliary variable,’ like AI-based Innovation, and ‘control auxil-
iary variables,’ like Time to Achieve SDG1 Targets, Inuence of SDG9,
and Poverty Headcount Ratio. However, the Poverty rate after taxes and
transfers can inadvertently increase poverty, slowing SDG1 progress.
In the SDG2 stock-ow chart (Fig. 8), the ‘stock’ or SDG2 score sig-
nies global hunger status. The ‘inow’ (rise in SDG2 score) reects
anti-hunger actions, while the ‘outow’ (fall in score) indicates
increasing hunger. AI-based Innovation (‘main auxiliary variable’) and
‘control auxiliary variables’, such as estimated SDG2 target achievement
year, SDG9 inuence, Human Trophic Level, Cereal Yield, Yield Gap
Closure, and Sustainable Nitrogen Management Index, shape the inow.
Conversely, factors like undernourishment prevalence, child stunting
and wasting, obesity, and hazardous pesticide exports boost the outow,
hindering SDG2 progress.
Fig. 6. Causal-loop diagram assessing the impact of AI-based Innovation on the SDGs.
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
11
Fig. 7. Stock-ow diagram assessing the impact of AI-based Innovation on SDG1 (Economic Pillar).
Fig. 8. Stock-ow diagram assessing the impact of AI-based Innovation on SDG2 (Economic Pillar).
Fig. 9. Stock-ow diagram assessing the impact of AI-based Innovation on SDG3 (Economic Pillar).
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
12
In the SDG3 stock-ow graph (see Fig. 9), the ‘stock’ is the SDG3
score, reecting global health. The scores for ‘inow’ and ‘outow’
indicate health improvements and hindrances, respectively. The ‘main
auxiliary variable’ is AI-based Innovation, while ‘control auxiliary var-
iables,’ such as the estimated SDG3 target achievement year, SDG9 in-
uence, life expectancy, adolescent fertility rate, birth assistance by
skilled health personnel, subjective well-being, and universal health
coverage, foster inow. Contrarily, factors like maternal and neonatal
mortality, tuberculosis incidence, pollution-related deaths, smoking
rates, disease death rates, new HIV infections, and trafc deaths fuel the
outow, suggesting ongoing health challenges.
In the stock-ow diagram for SDG4 (as shown in Fig. 10), the ‘stock’
is the SDG4 score, symbolizing global education quality. The
‘inow’—increased SDG4 score—and ‘outow’—decreased SDG4
score—represent educational improvements and challenges, respec-
tively. Inuencing the inow are the ‘main auxiliary variables,’ such as
AI-based Innovation, and ‘control auxiliary variables,’ such as Year to
Achieve SDG4 Targets, the inuence of SDG9 on SDG4 Targets, tech-
nology’s role in education, pre-primary organized learning
participation, net primary enrollment, literacy rate, and tertiary
attainment. However, factors such as low secondary completion and
science underachievers increase the SDG4 outow, highlighting
educational challenges.
In Fig. 11’s SDG5 stock-ow diagram, the ‘stock’ is the SDG5 score,
representing global gender equality status. The inow (representing an
increased SDG5 score) signies gender equality progress, while the
outow (indicating a decreased SDG5 score) denotes setbacks. The main
auxiliary variable, such as AI-based Innovation, and control auxiliary
variables, such as Year to Achieve SDG5 Targets, the inuence of SDG9
on SDG5 Targets, family planning methods, female-to-male education &
workforce ratio, and female parliamentary representation, drive the
inow. However, factors like the gender wage gap fuel the outow,
highlighting a key area needing policy attention for gender parity.
Fig. 12’s SDG6 stock-ow model uses the ‘stock’ as the SDG6 score to
denote the current global status of water and sanitation facilities. This
model features two pivotal streams: the ‘inow,’ representing im-
provements in water and sanitation, and the ‘outow,’ signifying the
challenges being encountered. The inow is inuenced by the ‘main
Fig. 10. Stock-ow diagram assessing the impact of AI-based Innovation on SDG4 (Social Pillar).
Fig. 11. Stock-ow diagram assessing the impact of AI-based Innovation on SDG5 (Social Pillar).
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
13
Fig. 12. Stock-ow diagram assessing the impact of AI-based Innovation on SDG6 (Environmental Pillar).
Fig. 13. Stock-ow diagram evaluating the impact of AI-based Innovation on SDG7 (Environmental Pillar).
Fig. 14. Stock-ow diagram assessing the impact of AI-based Innovation on SDG8 (Economic Pillar).
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
14
auxiliary variable,’ identied as AI-based Innovation, alongside ‘control
auxiliary variables’, such as Time to Achieve SDG6 Targets, how SDG6
Targets are inuenced by SDG9, accessibility of basic drinking water to
the population, effective and safely managed sanitation services, rate of
freshwater withdrawal, extent of wastewater treatment, and the amount
of scarce water consumption embodied within imports.
Fig. 15. Stock-ow diagram assessing the impact of AI-based Innovation on SDG10 (Social Pillar).
Fig. 16. Stock-ow diagram assessing the impact of AI-based Innovation on SDG11 (Social Pillar).
Fig. 17. Stock-ow diagram assessing the impact of AI-based Innovation on SDG12 (Environmental Pillar).
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
15
Fig. 18. Stock-ow diagram assessing the impact of AI-based Innovation on SDG13 (Environmental Pillar).
Fig. 19. Stock-ow diagram assessing the impact of AI-based Innovation on SDG14 (Environmental Pillar).
Fig. 20. Stock-ow diagram assessing the impact of AI-based Innovation on SDG15 (Environmental Pillar).
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
16
For SDG7, as depicted in Fig. 13, the ‘stock’ represents the SDG7
score, highlighting the current global situation of Affordable & Clean
Energy. The ‘inows’ reect improvements in the score, arising from
advances in affordable, clean energy, while ‘outows’ represent de-
creases due to challenges like rising CO
2
emissions from fuel combus-
tion. The inow is inuenced by ‘main auxiliary variables,’ like AI-based
Innovation, and ‘control auxiliary variables’, such as the projected Time
to Achieve SDG7 Targets, the inuence of SDG9 on the SDG7 Target, the
percentage of people with access to electricity and clean cooking fuels,
as well as the share of renewable energy in the total primary energy
supply, all of which strengthen the SDG7 inow.
In the stock-ow illustration pertaining to SDG8, as presented in
Fig. 14, the SDG8 score serves as the ‘stock’ representing the present
international status of quality employment and economic growth. The
‘inows’ symbolize an upward trajectory in the SDG8 score, thereby
encapsulating progress in aspects such as labor rights, accessibility to
banking services, and adjusted growth in gross domestic product. A
‘main auxiliary variable’ controlling the pace of inow encompasses ‘AI-
based Innovation,’ whereas the rate of inow is modulated by ‘control
auxiliary variables’ like the Year to Achieve SDG8 Targets and SDG8
Target Inuenced by SDG9. On the other hand, the ‘outows’ denote a
decline in the SDG8 score, attributed to negative factors like elevated
youth unemployment, pervasive modern slavery, substantial ination
rate, widespread unemployment, and incidence of work-related
fatalities.
In the stock-ow schematic concerning SDG10, as depicted in
Fig. 15, the SDG10 score embodies the current worldwide status of
equality, thereby constituting the ‘stock.’ The ‘inows’ correspond to an
upsurge in the SDG10 score, representing noteworthy advances towards
absolute parity in income and wealth. Conversely, the ‘outows’ denote
a diminution in the SDG10 score, illustrating worsening disparities
primarily attributed to uneven income distribution and poverty rates
among the elderly population. The primary auxiliary variable regulating
the inow rate encompasses AI-based innovation, while the control
auxiliary variable, which includes Year to Achieve SDG10 Targets and
SDG10 Target Inuenced by SDG9, modulates the inow rate.
In the stock-ow diagram for SDG11 (Fig. 16), the ‘stock’ is the
SDG11 score, representing the current global state of urban sustain-
ability. An increase in the SDG11 score represents the ‘inow,’ signi-
fying progress in urban sustainability. The rate of this inow is regulated
by the ‘main auxiliary variable,’ AI-based innovation, and ‘control
auxiliary variables’ including Year to Achieve SDG11 Targets, SDG11
Target Inuenced by SDG9, access to improved water sources for urban
dwellers, and satisfaction with public transportation systems.
Conversely, the ‘outow’ signies a decrease in the SDG11 score,
highlighting challenges to urban sustainability, such as widespread
slums, particulate matter pollution, and high rent burdens.
Within the stock-ow framework for SDG12, as depicted in Fig. 17,
the SDG12 score serves as the ‘stock,’ signifying the prevailing global
condition of sustainable consumption and production. An upsurge in the
SDG12 score constitutes the ‘inow,’ representing benecial transitions
towards conscientious consumption. The rate of this inow is regulated
by the ‘primary auxiliary variable,’ such as AI-based innovation, and
‘control auxiliary variables’, such as Year to Achieve SDG12 Targets and
SDG12 Target Inuenced by SDG9. In contrast, the ‘outow’ intimates a
diminution in the SDG12 score, highlighting problems such as excessive
Fig. 21. Stock-ow diagram assessing the impact of AI-based Innovation on SDG16 (Social Pillar).
Fig. 22. Stock-ow diagram assessing the impact of AI-based Innovation on SDG17 (Social Pillar).
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
17
waste generation and emissions, which have detrimental effects on
sustainable practices. Control auxiliary variables contributing to this
‘outow’ include elements like municipal solid waste, electronic waste,
SO
2,
and nitrogen emissions, exports of plastic waste, and non-recycled
municipal solid waste.
In the SDG13 stock-ow framework, as delineated in Fig. 18, the
‘stock’ is denoted by the SDG13 score, reecting the current global
advancement in climate action. The ‘inow’ corresponds to a rise in the
SDG13 score, which is inuenced by AI-based innovation as the ‘primary
auxiliary variable’ and ‘control auxiliary variables,’ such as Year to
Achieve SDG13 Targets, SDG13 Target Inuenced by SDG9, and an
elevated Carbon Pricing Score, which symbolizes efcacious mitigation
approaches. Conversely, the ‘outow’ signies a reduction in the SDG13
score, attributable to factors such as CO
2
emissions from fossil fuel
combustion, cement manufacturing, imports, and fossil fuel exports,
thereby highlighting the environmental burden.
In the SDG14 stock-ow diagram (see Fig. 19), the ‘stock’ is repre-
sented by the SDG14 score, denoting the health of marine ecosystems.
An increase in the SDG14 score signies the ‘inow,’ driven by AI-based
innovation as the ‘primary auxiliary variable.’ Other factors inuencing
the inow include ‘control auxiliary variables,’ such as the Year to
Achieve SDG14 Targets, SDG14 Target Inuenced by SDG9, the average
area of protection in crucial marine biodiversity sites, and the Ocean
Health Index: Clean Waters score, highlighting successes in marine
conservation. Conversely, the ‘outow,’ indicating a decrease in the
SDG14 score, arises from issues like sh extraction from overstrained or
depleted stocks, trawling, discarding, and import-related threats to
marine biodiversity.
In the stock-ow model pertaining to SDG15, as illustrated in Fig. 20,
the SDG15 score operates as the ‘stock,’ portraying the health of
terrestrial and freshwater biodiversity and ecosystems. The ‘inow,’
symbolizing an enhancement in the SDG15 score, is inuenced by AI-
based innovation as the ‘main auxiliary variable,’ and Year to Achieve
SDG15 Targets, SDG15 Target Inuenced by SDG9, Mean Area Protected
in Terrestrial and Freshwater Sites Important to Biodiversity, as ‘control
auxiliary variables’ that bolster the inow rate of SDG15, signifying
advances in land and freshwater conservation. Contrarily, the ‘outow,’
representing a diminution in the SDG15 score, is inuenced by factors
such as the Red List Index of Species Survival, Permanent Deforestation,
and Terrestrial and Freshwater Biodiversity Threats Embodied in
Imports.
Within the stock-ow model for SDG16, as delineated in Fig. 21, the
SDG16 score constitutes the ‘stock,’ reecting the global standing of
peace, justice, and robust institutions. The ‘inow’ into this stock, a rise
in the SDG16 score, is mediated by variables including AI-based inno-
vation as the ‘primary auxiliary variable,’ and Year to Achieve SDG16
Targets, SDG16 Target Inuenced by SDG9, the population’s perception
of safety, property rights, birth registrations, access to justice, and press
freedom, as the ‘control auxiliary variables.’ In contrast, the ‘outow,’
which signies a reduction in the SDG16 score, is shaped by variables
such as the prevalence of homicides, the number of unsentenced de-
tainees, child labor involvement, levels of corruption, exports of con-
ventional weaponry, and the number of incarcerated individuals.
In a stock-ow diagram for SDG17 (see Fig. 22, below), the ‘stock’ is
the SDG17 score (Partnerships for the Goals), reecting the global
progress in fostering partnerships for achieving the SDGs. The ‘inow,’
signifying an increase in the SDG17 score, is driven by elements such as
AI-based innovation as the ‘main auxiliary variable,’ and Year to Ach-
ieve SDG17 Targets, SDG17 Target inuenced by SDG9, international
concessional public nance for high-income and all OECD DAC coun-
tries, government revenue (excluding grants), the state of corporate tax
havens, and government spending on health and education, as the
‘control auxiliary variables.’ Conversely, the ‘outow’ represents a
decrease in the SDG17 score, dictated by elements like nancial secrecy
and shifted prots of multinationals, which expedite the SDG17 outow
rate.
3.10. Model validation procedure
Before the study’s simulation, a rigorous validation process was
undertaken, following Wu et al. (2010). Initial steps included con-
ducting a ‘dimensional consistency’ test, involving cross-verifying the
model’s dimensions dened in Table 1 with real-world counterparts,
conrmed by experts. In addition, a ‘structural verication’ test was
performed using the ‘Check Model’ function in Vensim software, veri-
fying the model’s accuracy in terms of variable relations and formula-
tions. A nal ‘behavioral validation’ test compared simulation data
against historical data from 2021 to 2022, encompassing key variables
such as SDG achievement and AI-based innovation. With deviations <5
%, the model’s accuracy was substantiated, thus proving the simula-
tion’s reliability for future forecasting.
4. Results and analysis
The simulation of the model is carried out over nine years, beginning
in 2022 and ending in 2030. The result for each SDG is provided below:
4.1. SDG1 (ending poverty in all its forms worldwide) related results
SDG1 (Economic Pillar) aims to end global poverty. Seven hundred
and sixty-seven million developing country residents—1 in 5—live on
less than $1.90 per day. Southern Asia and sub-Saharan Africa suffer the
most. Political-economic instability is linked to extreme poverty (United
Nations, 2022d). Table 2 in the Appendix presents the results for the
impact of AI-based innovation on SDG1.
From Table 2, it is evident that the estimated impact of AI-based
innovation on SDG1 (No Poverty) is to decrease signicantly in 2022,
compared to the base year of 2021, and then increase continuously
between 2023 and 2030 for developed nations such as Australia, Can-
ada, France, Germany, Japan, New Zealand, Turkey, Russia, the UAE,
the United Kingdom, and the United States. The same applies to devel-
oping nations such as Pakistan, Nigeria, Egypt, India, Vietnam,
Bangladesh, and Iran.
In Saudi Arabia (a developed country), the impact of AI-based
innovation on SDG1 remained unchanged between 2021 and 2030.
Moreover, China and Kenya are estimated to show decrease-increase-
decrease and decrease-increase-decrease-increase trends, respectively.
4.2. SDG2 (zero hunger) related results
SDG2 (Economic Pillar) seeks global hunger eradication by 2030. In
2020, 161 million more people were hungrier than in 2019. Hunger and
food insecurity increased between 2014 and the COVID-19 pandemic.
COVID-19 has exacerbated malnutrition, especially in children. The
Ukraine war has recently caused a severe food crisis, disrupting global
supply chains since the Second World War (United Nations, 2022d).
Table 3 in the Appendix shows how AI-based innovation has affected
SDG2.
From Table 3, it is evident that the impact of AI-based innovation on
SDG2 (Zero Hunger) is anticipated to decrease signicantly in 2022,
compared to the base year of 2021, and then increase continuously
between 2023 and 2030 for developed nations such as Australia, Ger-
many, Japan, Russia, Saudi Arabia, and the UAE. The same applies to
developing countries such as Vietnam, Uzbekistan, Pakistan, Nigeria,
Kenya, Egypt, Bangladesh, China, and India.
For the developed countries France, New Zealand, Turkey, and the
UK, the inuence of AI-based innovation on SDG2 is projected to fall
continuously between 2021 and 2030. The same applies to the devel-
oping country Iran. For the USA, the inuence of AI-based innovation on
SDG2 is projected to fall continuously between 2021 and 2026. Then,
the USA is estimated to witness a steady increase between 2027 and
2029.
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
18
4.3. SDG3 (good health and well-being for all) related results
Sustainable development requires healthy lives and all-age well-
being (SDG3). Unfortunately, COVID-19 has caused human suffering,
destabilized the global economy, and disrupted the progress made in
health (e.g., improved life expectancy and child and maternal mortality)
before the pandemic. Hence, more efforts are needed to address many
persistent and emerging health issues through health system funding,
sanitation, hygiene, and physician access (United Nations, 2022d).
Table 4 in the Appendix summarizes the impact of AI-based innovation
on SDG3 (Economic Pillar).
From Table 4, it is evident that the impact of AI-based innovation on
SDG3 (Good Health and Well-Being) is anticipated to decrease signi-
cantly in 2022, compared to the base year of 2021, and then increase
continuously between 2023 and 2030 for developed nations such as
Australia, New Zealand, France, Germany, Japan, Russia, Saudi Arabia,
the UAE, the USA, and the UK. The same applies to developing countries
such as Vietnam, Uzbekistan, Pakistan, Nigeria, Kenya, Egypt,
Bangladesh, China, Iran, and India.
For the developed countries, Canada and Turkey, the inuence of AI-
based innovation on SDG3 is projected to signicantly decrease in 2022,
compared to the base year of 2021, and then steadily increase between
2023 and 2024. Then, it is projected to drop signicantly in 2025 before
continuously increasing from 2026 to 2030.
4.4. SDG4 (quality education) related results
SDG4 (Social Pillar) stands for quality education. Over the past
decade, school enrollment, especially for girls, has increased. Nonethe-
less, in 2018, nearly 20 % of the world’s youth were out of school. The
global pandemic has also threatened global education progress (United
Nations, 2022d). Table 5 in the Appendix shows how AI-based innova-
tion affected SDG4.
The impact of AI-based innovation on SDG4 is anticipated to
decrease signicantly in 2022, compared to the base year of 2021, and
then increase continuously between 2023 and 2030 for developed na-
tions such as Australia, Canada, New Zealand, Russia, Saudi Arabia, and
the UAE. The same applies to developing countries such as Vietnam,
Uzbekistan, Egypt, Bangladesh, and Iran.
For the developed country, the USA, the inuence of AI-based
innovation on SDG4 is projected to drop continuously between 2021
and 2030. On the other hand, France, Japan, China, and Kenya are
projected to show an increase-decrease trend. Moreover, the USA,
Turkey, India, Pakistan, and Nigeria are expected to show a decrease-
increase-decrease trend, while Germany will have a decrease-increase-
decrease-increase trend.
4.5. SDG5 (gender equality) related results
Gender equality is required for stability, prosperity, and sustainable
development. In the last few decades, there has been a rise in the edu-
cation of young women, a decline in the practice of early marriage, an
increase in the number of powerful women in positions of authority, and
the passing of legislation to further advance gender equality. There are
still many obstacles to overcome, despite these advances. In 2019, one in
every ve women and girls between the ages of 15 and 49 experienced
physical or sexual abuse at the hands of an intimate partner, discrimi-
natory legislation and societal norms persisted, and women were un-
derrepresented at every level of government. COVID-19 further
exacerbated gender disparities in terms of security, the economy, and
social protection (United Nations, 2022c). Table 6 in the Appendix
summarizes the impact of AI-based innovation on SDG5.
From Table 6, it is evident that the impact of AI-based innovation on
SDG5 (Gender Equality) is anticipated to increase continuously between
2021 and 2030 for the developed nations of Japan and Russia. The same
applies to developing countries such as China and Pakistan.
For developed nations such as Australia, the USA, the UK, the UAE,
Canada, France, New Zealand, Saudi Arabia, and Turkey, the impact of
AI-based innovation on SDG5 is anticipated to decrease signicantly in
2022, compared to the base year of 2021, and then increase continu-
ously between 2023 and 2030. The same applies to developing countries
such as India, Bangladesh, Kenya, Egypt, and Iran.
Nigeria (a developing country), Uzbekistan (a developing country),
and Germany (a developed country) are projected to witness Decrease-
Increase, Decrease-Increase-Decrease-Increase, and Increase-Decrease-
Increase trends, respectively.
4.6. SDG6 (clean water and sanitation) related results
Despite progress, a billion people—mostly rural—lack clean water
and sanitation. One in three people lack safe drinking water, two in ve
lack soap and water for handwashing, and over 673 million defecate in
the open (United Nations, 2022b). Table 7 in the Appendix presents the
results of the impact of AI-based innovation on SDG6 (Environmental
Pillar).
From Table 7, it is evident that the impact of AI-based innovation on
SDG6 (Clean Water and Sanitation) is anticipated to increase continu-
ously between 2021 and 2030 for the developed nations of Australia,
France, Germany, Japan, Russia, Saudi Arabia, and the UAE. The same
applies to developing countries such as Egypt, Iran, Nigeria, and
Pakistan. On the other hand, the impact of AI-based innovation on SDG6
is anticipated to decrease continuously between 2021 and 2030 for the
developed nation of Canada. The same applies to the developing country
Uzbekistan.
For developing nations such as Vietnam, Kenya, Bangladesh, China,
and India, the impact of AI-based innovation on SDG6 is anticipated to
decrease signicantly in 2022, compared to the base year of 2021, and
then increase continuously between 2023 and 2030. On the other hand,
for developed nations such as the UK and Turkey, the impact of AI-based
innovation on SDG6 is anticipated to increase signicantly in 2022,
compared to the base year of 2021, and then decrease continuously
between 2023 and 2030.
In New Zealand (a developed country), the impact of AI-based
innovation on SDG6 is forecast to remain unchanged between 2021
and 2030.
In the USA (a developed country), the inuence of AI-based inno-
vation on SDG6 is projected to see a steady increase from 2022 to 2027,
compared to the base year of 2021, then a slight drop in 2028, followed
by another steady increase from 2029 to 2030.
4.7. SDG7 (affordable and clean energy) related results
By improving energy efciency, renewable energy, and electricity
for developing countries, Goal 7 is being accomplished. Nonetheless,
more effort is required to electrify Sub-Saharan Africa, boost access to
clean and safe cooking methods and fuels for the world’s 3 billion people
who lack them, and diversify renewable energy sources beyond elec-
tricity (United Nations, 2022c). Table 8 in the Appendix shows how AI-
based innovation affects SDG7 (Environmental Pillar).
Table 8 shows that the impact of AI-based innovation on SDG7
(Affordable and Clean Energy) is anticipated to increase continuously
between 2021 and 2030 for the developed nations of France, New
Zealand, Saudi Arabia, and the USA. On the other hand, the impact of AI-
based innovation on SDG7 is anticipated to decrease continuously be-
tween 2021 and 2030 for the developing nation of Egypt.
For developing nations such as Uzbekistan, Vietnam, Kenya,
Bangladesh, China, and India, the impact of AI-based innovation on
SDG7 is anticipated to decrease signicantly in 2022, compared to the
base year of 2021, and then increase continuously between 2023 and
2030. The same is true for developed nations such as the UK, Australia,
Japan, Germany, and Turkey. On the other hand, for Russia (a developed
country), the impact of AI-based innovation on SDG7 is anticipated to
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
19
increase signicantly in 2022, compared to the base year of 2021, and
then decrease continuously between 2023 and 2030.
Pakistan (a developing country) and Canada (a developed country)
are projected to witness an Increase-Decrease trend. On the other hand,
it is anticipated that the UAE, Iran, and Nigeria will witness an increase-
decrease-increase trend.
4.8. SDG8 (decent work and economic growth) related results
A global recession worse than in 2009 in the aftermath of COVID-19
is forecast by the International Monetary Fund (IMF). The ILO estimates
that nearly half of the global workforce may lose their jobs (United
Nations, 2022c). Under this circumstance, inclusive economic growth is
needed to boost living standards, jobs, and progress. Table 9 in the
Appendix shows how AI-based innovation affects SDG8 (Economic
Pillar).
From Table 9, it is evident that the impact of AI-based innovation on
SDG8 (Decent Work and Economic Growth) is anticipated to increase
continuously between 2021 and 2030 for the developed nations of
France and New Zealand.
For developing nations such as Uzbekistan, Nigeria, Iran, Egypt,
Kenya, Bangladesh, and China, the impact of AI-based innovation on
SDG8 is anticipated to decrease signicantly in 2022, compared to the
base year of 2021, and then increase continuously between 2023 and
2030. The same is true for developed nations such as the UAE, Russia,
the UK, Germany, and Turkey. On the other hand, for Saudi Arabia (a
developed country), the impact of AI-based innovation on SDG8 is
anticipated to increase signicantly in 2022, compared to the base year
of 2021, and then decrease continuously between 2023 and 2030.
In Australia (a developed country), the inuence of AI-based inno-
vation on SDG8 is projected to see a steady increase from 2022 to 2026,
compared to the base year of 2021, followed by a steep drop in 2027 and
another steady increase from 2028 to 2030. On the other hand, it is
anticipated that the USA, Japan, Canada, and Vietnam will witness a
Decrease-Increase-Decrease-Increase trend. Similarly, India and
Pakistan are forecast to witness Decrease-Increase-Decrease and
Decrease-Increase trends, respectively.
4.9. SDG10 (reduced inequalities) related results
Even though some countries have lower relative income inequality
and lower-income countries have preferential trade status, inequality
persists. COVID-19 has worsened inequalities, especially for the poor
and vulnerable, highlighting economic inequality and fragile social
safety nets (United Nations, 2022d). Table 10 in the Appendix presents
the results for the impact of AI-based innovation on SDG10 (Social
Pillar).
From Table 10, it is evident that the impact of AI-based innovation on
SDG10 (Reduced Inequalities) is anticipated to increase continuously
between 2021 and 2030 for the developed nation of Japan. On the other
hand, the impact of AI-based innovation on SDG10 is anticipated to
decrease continuously between 2021 and 2030 for Australia, Germany,
New Zealand, Egypt, and India.
For developing nations such as Nigeria, Kenya, and China, the impact
of AI-based innovation on SDG10 is anticipated to decrease signicantly
in 2022, compared to the base year of 2021, and then increase contin-
uously between 2023 and 2030. The same is true for developed nations
such as France, Canada, Russia, and the UAE. On the other hand, for
Turkey, Bangladesh, and Iran, the impact of AI-based innovation on
SDG10 is anticipated to increase signicantly in 2022, compared to the
base year of 2021, and then decrease continuously between 2023 and
2030.
In Saudi Arabia (a developed country), the impact of AI-based
innovation on SDG10 is forecast to remain unchanged between 2021
and 2030.
It is anticipated that the USA and the UK will witness an ‘Increase-
Decrease-Increase- Decrease’ trend. On the other hand, Uzbekistan and
Pakistan are forecast to witness a Decrease-Increase-Decrease trend.
Similarly, Vietnam is forecast to witness a ‘Decrease-Increase’ trend.
4.10. SDG11 (sustainable cities and communities) related results
Cities and metropolitan areas drive 60 % of global GDP and produce
70 % of carbon emissions. Rapid urbanization is leading to more slum
dwellers, overburdened infrastructure and services, and worsening air
pollution. The Food and Agricultural Organization (FAO) warns that
without measures to feed poor and vulnerable urban residents, hunger
and fatalities could rise signicantly (United Nations, 2022d). Table 11
in the Appendix presents the results for the impact of AI-based innova-
tion on SDG11 (Social Pillar).
Table 11 shows that the impact of AI-based innovation on SDG11
(Sustainable Cities and Communities) is anticipated to increase contin-
uously between 2021 and 2030 for Canada, the USA, and India. On the
other hand, the impact of AI-based innovation on SDG11 is anticipated
to decrease continuously between 2021 and 2030 for the developing
nation of Uzbekistan.
For developing nations such as Vietnam, Bangladesh, Pakistan,
Kenya, and China, the impact of AI-based innovation on SDG11 is
anticipated to decrease signicantly in 2022, compared to the base year
of 2021, and then increase continuously between 2023 and 2030. The
same is true for developed nations such as Australia, New Zealand,
Russia, and the UAE. On the other hand, for the UK, Germany, Turkey,
Nigeria, and Iran, the impact of AI-based innovation on SDG11 is
anticipated to face an ‘increase-decrease’ trend, whereas Japan and
Saudi Arabia are forecast to show a ‘decrease-increase’ trend, Egypt a
‘decrease-increase-decrease-increase’ trend, and France a ‘decrease-in-
crease-decrease’ trend.
4.11. SDG12 (responsible consumption and production) related results
Incessant consumption and production continue to harm the envi-
ronment. In the past 100 years, economic and social progress has gone
hand in hand with environmental damage, putting at risk the systems on
which our future growth and survival depend (United Nations, 2022d).
Table 12 in the Appendix shows how AI-based innovation affects SDG12
(Environmental Pillar).
Table 12 shows that the impact of AI-based innovation on SDG12
(Responsible Production and Consumption) is anticipated to decrease
continuously between 2021 and 2030 for Australia, Canada, Germany,
Japan, New Zealand, Egypt, China, Saudi Arabia, and Uzbekistan.
For the developing nation of Vietnam, the impact of AI-based inno-
vation on SDG12 is anticipated to decrease signicantly in 2022,
compared to the base year of 2021, and then increase continuously
between 2023 and 2030. On the other hand, for France, Turkey, the
USA, the UK, the UAE, India, Bangladesh, Kenya, Nigeria, Pakistan,
Russia, and Iran, the impact of AI-based innovation on SDG12 is antic-
ipated to increase signicantly in 2022, compared to the base year of
2021, and then decrease continuously between 2023 and 2030.
4.12. SDG13 (actions to combat climate change) related results
The 2015 Paris Agreement set a limit of 2 ◦C for the increase in global
temperature this century. Through nancial ows, new technology
frameworks, and improved capacity building, the agreement also aims
to help countries cope with climate change (United Nations, 2022c).
Table 13 in the Appendix shows how AI-based innovation affects SDG13
(Environmental Pillar).
Table 13 shows that the impact of AI-based innovation on SDG13
(Climate Action) is anticipated to increase continuously between 2021
and 2030 for France and Nigeria. On the other hand, the impact of AI-
based innovation on SDG13 is anticipated to decrease continuously
between 2021 and 2030 for Uzbekistan, China, India, Kenya, and
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
20
Pakistan.
For Australia, Germany, Japan, New Zealand, Canada, Turkey, the
UAE, the UK, the USA, Saudi Arabia, Egypt, Vietnam, and Russia, the
impact of AI-based innovation on SDG13 is anticipated to decrease
signicantly in 2022, compared to the base year of 2021, and then in-
crease continuously between 2023 and 2030. On the other hand, for
Bangladesh, the impact of AI-based innovation on SDG13 is anticipated
to increase signicantly in 2022, compared to the base year of 2021, and
then decrease continuously between 2023 and 2030.
For Iran, the impact of AI-based innovation on SDG13 is anticipated
to face a ‘decrease-increase’ trend.
4.13. SDG14 (protecting and promoting sustainability for life below
water) related results
Sustainable futures require careful management of oceans and seas,
which deliver and regulate our drinking water, rainwater, climate,
weather, food, coastlines, and oxygen. Pollution degrades coastal wa-
ters, and ocean acidication harms ecosystems and biodiversity. Ocean
conservation must continue by reducing overshing and marine pollu-
tion (United Nations, 2022c). Table 14 in the Appendix summarizes the
impact of AI-based innovation on SDG14 (Environmental Pillar).
Table 14 shows that the impact of AI-based innovation on SDG14
(Life Below Water) is anticipated to increase continuously between 2021
and 2030 for Germany, Iran, and Pakistan. On the other hand, the
impact of AI-based innovation on SDG14 is anticipated to decrease
continuously between 2021 and 2030 for Uzbekistan, China, India,
Nigeria, Saudi Arabia, and Turkey.
For the UK, the USA, and France, the impact of AI-based innovation
on SDG14 is anticipated to decrease signicantly in 2022, compared to
the base year of 2021, and then increase continuously between 2023 and
2030. On the other hand, for Japan, Canada, and New Zealand, the
impact of AI-based innovation on SDG14 is anticipated to increase
signicantly in 2022, compared to the base year of 2021, and then
decrease continuously between 2023 and 2030.
For Russia, the UAE, and Bangladesh, the impact of AI-based inno-
vation on SDG14 is anticipated to face ‘increase-decrease-increase-
decrease-increase-decrease,’ ‘increase-decrease-increase-decrease-in-
crease,’ and ‘increase-decrease-increase-decrease,’ trends, respectively.
On the other hand, Egypt and Kenya are forecast to face a ‘decrease-
increase-decrease’ trend.
4.14. SDG15 (protecting life on land) related results
Humans have modied nearly all of Earth’s surface, leaving less and
less room for the planet’s natural inhabitants. It is estimated that one
million species of animals and plants will become extinct in the next
century. Radical shifts are required to preserve and revive the natural
world (United Nations, 2022d). The effects of AI-based innovation on
SDG15 (Environmental Pillar) are summarized in Table 15 in the
Appendix.
Table 15 shows that the impact of AI-based innovation on SDG15
(Life on Land) is anticipated to increase continuously between 2021 and
2030 for Australia, Canada, the UAE, the UK, and the USA. On the other
hand, the impact of AI-based innovation on SDG15 is anticipated to
decrease continuously between 2021 and 2030 for Egypt, Pakistan,
Russia, and Saudi Arabia.
For France and Uzbekistan, the impact of AI-based innovation on
SDG15 is anticipated to decrease signicantly in 2022, compared to the
base year of 2021, and increase continuously between 2023 and 2030.
On the other hand, for Bangladesh, China, India, Iran, Japan, Kenya,
Nigeria, Turkey, and New Zealand, the impact of AI-based innovation on
SDG15 is anticipated to increase signicantly in 2022, compared to the
base year of 2021, and then decrease continuously between 2023 and
2030.
For Germany and Vietnam, the impact of AI-based innovation on
SDG15 is anticipated to face ‘increase-decrease-increase-decrease-in-
crease,’ and ‘decrease-increase’ trends, respectively.
4.15. SDG16 (peace, justice, and strong institutions) related results
Conict, weak institutions, insecurity, and injustice threaten sus-
tainable development. The United Nations High Commissioner for Ref-
ugees (UNHCR) registered a record 70 million refugees in 2018. The UN
reported that 357 journalists, human rights defenders, and trade
unionists were murdered, and 30 went missing in 47 nations in 2019.
Currently, one in four under-5 children worldwide is unregistered,
denying them a legal identity and access to justice and social services
(United Nations, 2022b). Table 16 in the Appendix shows how AI-based
innovation affects SDG16 (Social Pillar).
Table 16 shows that the impact of AI-based innovation on SDG16
(Peace, Justice, and Solid Institutions) is anticipated to increase
continuously between 2021 and 2030 for Japan, Russia, and Turkey. On
the other hand, the impact of AI-based innovation on SDG16 is antici-
pated to decrease continuously between 2021 and 2030 for India.
For France, Iran, China, Egypt, Pakistan, Saudi Arabia, the UAE,
Vietnam, and Uzbekistan, the impact of AI-based innovation on SDG16
is anticipated to decrease signicantly in 2022, compared to the base
year of 2021, and increase continuously between 2023 and 2030. On the
other hand, for Bangladesh, the UK, Canada, the USA, Germany, and
New Zealand, the impact of AI-based innovation on SDG16 is anticipated
to increase signicantly in 2022, compared to the base year of 2021, and
then decrease continuously between 2023 and 2030. Similarly, Kenya
and Nigeria are forecast to face a ‘decrease-increase-decrease’ trend.
For Australia, the impact of AI-based innovation on SDG16 is
anticipated to face an ‘increase-decrease-increase-decrease’ trend.
4.16. SDG17 (revitalize the global partnership for sustainable
development) related results
The COVID-19 pandemic might have caused the worst recession
since the Great Depression, a 3 % global economic contraction in 2020.
SDG achievement requires strong international cooperation. SDG17
prefers AI-driven connections for problem-solving and SDG goals
(United Nations, 2022a). Table 17 in the Appendix shows how AI-based
innovation affects SDG17 (Social Pillar).
Table 17 indicates that the impact of AI-based innovation on SDG17
(Partnerships for the Goals) is anticipated to decrease continuously be-
tween 2021 and 2030 for Egypt. On the other hand, the impact of AI-
based innovation on SDG17 is anticipated to increase continuously be-
tween 2021 and 2030 in New Zealand.
For Australia, Canada, France, Germany, India, Iran, China, Japan,
Turkey, Nigeria, Pakistan, Vietnam, and Uzbekistan, the impact of AI-
based innovation on SDG17 is anticipated to decrease signicantly in
2022, compared to the base year of 2021, and increase continuously
between 2023 and 2030. On the other hand, for Bangladesh and Kenya,
the impact of AI-based innovation on SDG17 is anticipated to increase
signicantly in 2022, compared to the base year of 2021, and then
decrease continuously between 2023 and 2030. Similarly, the UK and
the UAE are forecast to face an ‘increase-decrease’ trend.
For the USA and Saudi Arabia, the impact of AI-based innovation on
SDG17 is anticipated to face ‘increase-decrease-increase-decrease’ and
‘increase-decrease-increase,’ trends, respectively. On the other hand, for
Russia, the impact of AI-based innovation on SDG17 is anticipated to
face a ‘decrease-increase-decrease-increase’ trend.
5. Findings and discussion
5.1. Main ndings
The widespread consensus is that AI and associated technologies,
such as machine intelligence and big data, have revolutionized our
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
21
present and future (Demiralay et al., 2021). According to numerous
previous studies, AI is linked to education, employment, health care,
utilities, and environmental protection (Amabile, 2020; Bag et al., 2021;
Füller et al., 2022). However, there are also concerns because, as
anticipated, AI might outperform humans. Taking into account both
these positive and negative aspects of AI (Lee et al., 2022), I created a
model based on TEF (theoretical framework) to predict the effects of AI-
based Innovation on SDGs using a System Dynamics Perspective in a
cross-country setting. The results of hypotheses tests based on this model
are summarized in Table 18 in the Appendix. When most of the 22
countries (except for a few) supported a hypothesis, a hypothesis is
considered to be ‘supported.’ On the other hand, when only half of the
countries supported a hypothesis, a hypothesis is considered to be
‘partially supported.’ Otherwise, a hypothesis is not supported.
First, since 2022 was affected by COVID-19, I have considered it an
exceptional case for all the SDGs’ achievements. Goodell (2020) sup-
ports this by arguing that the pandemic has sped up the adoption of AI
and associated technologies in healthcare, education, and remote work,
potentially alleviating poverty. Hence, I can conclude that H1a, which
proposed that AI-based innovation (SDG9) positively inuences SDG1
(poverty reduction), is overall supported for both developed and
developing countries. Countries are likely to perform better on SDG1 if
they utilize free digital platforms for nancial management, Accounting
Information Systems for nancial education, and AI for economic data
analysis, challenging populist views that charity and wealth redistri-
bution alone can alleviate poverty (Keding and Meissner, 2021).
Nevertheless, developed countries are expected to perform better in
achieving SDG1 than developing ones. Hence, H4a, which proposed that
AI-based innovation helps developed countries achieve better results in
SDG1 due to better institutional arrangements than developing coun-
tries, is overall supported. This is consistent with the ndings of Benzidia
et al. (2021), who reported that AI-based innovation can mitigate
poverty by boosting economic growth, generating jobs, and improving
social welfare.
Second, H1b, which proposed that AI-based innovation (SDG9) im-
pacts positively on SDG2 (zero hunger), is supported by some developed
(Australia, Germany, Japan, Russia, Saudi Arabia, and the UAE) and
developing countries (e.g., Vietnam, Uzbekistan, Pakistan, Nigeria,
Kenya, Egypt, Bangladesh, China, and India) based on 2023–2030
forecasting. The predicted success of these nations is due to regulations
like food governance-related Agricultural Innovation Systems, healthy
eating education, and risk warning systems (Keding and Meissner,
2021), aligning with the ndings of Elsayed et al. (2020) ndings on AI’s
potential to enhance agricultural efciency, safety, and productivity.
Nonetheless, H1b is not supported for other developed (e.g., France,
New Zealand, Turkey, USA, and UK) and developing countries (Iran),
emphasizing the role of context-specic factors, such as access to
affordable labor (Gupta et al., 2020). Hence, H1b is partially supported
overall, and developing countries are anticipated to show better per-
formance
20
in achieving SDG2 than developed countries. Hence, H4b,
which proposed that AI-based innovation helps developed countries to
achieve better results in SDG2 due to better institutional arrangements
than developing countries, is not supported. This nding aligns with
Zameer et al. (2020), indicating that factors beyond institutional ar-
rangements, such as resourceful innovative practices and comparative
advantages like affordable labor, can also enable developing countries to
achieve zero hunger.
Third, H1c, indicating that AI-based innovation (SDG9) positively
inuences SDG3 (Good Health and Well-being), is corroborated by the
estimated results for 2023–2030 for most developed and all developing
countries. These countries leverage AI in medical prediction and diag-
nosis, healthcare training through VR, AR, and mobile apps, and use
public digital apps to promote healthy habits and control infectious
diseases, in line with the ndings of Benzidia et al. (2021). However,
H1c is not supported in developed countries, Canada and Turkey. Hence,
H1c is supported
21
overall, and developing countries are anticipated to
show better performance
22
in achieving SDG3 compared to developed
countries. Hence, H4c, which proposed that better institutional ar-
rangements in developed countries help them perform better in
achieving AI-based innovation and resultant better SDG3, is not wholly
supported. This is consistent with the nding of Li et al. (2021) that AI
can improve healthcare in the resource-limited settings typically found
in developing countries.
Fourth, H2a, which proposed that AI-based innovation (SDG9)
positively inuences SDG4 (quality education), is supported by the
estimated results for 2023–2030 for some developed (e.g., Australia,
Canada, New Zealand, Russia, Saudi Arabia, and the UAE) and devel-
oping countries (e.g., Vietnam, Uzbekistan, Egypt, Bangladesh, and
Iran). These countries employ advanced digital platforms, AI, and big
data to enhance education access and quality, fostering personalized and
inclusive learning, as supported by research illustrating AI’s positive
inuence on learning outcomes (Benzidia et al., 2021; Grijalvo et al.,
2022). Yet, H2a is not supported by other developed (e.g., Japan, Ger-
many, Turkey, the USA, and France) and developing countries (Nigeria,
China, India, Kenya, and Pakistan). Hence, H2a is partially supported.
This aligns with Calabrese et al. (2023), who found that AI-based edu-
cation’s effectiveness may depend on specic cultural and social con-
texts, as well as policy frameworks. These results show that developed
and developing countries are expected to perform similarly in achieving
SDG4. Hence, H4d, which proposed that AI-based innovation helps
developed countries to achieve better results in SDG4 due to better
institutional arrangements than developing countries, is not supported.
This contradicts the research of Füller et al. (2022), arguing that
developing countries might struggle to adopt and implement AI-based
innovations due to weaker institutional capacities.
Fifth, H2b, which proposed that AI-based innovation (SDG9) posi-
tively inuences SDG5 (gender equality), is supported by the estimated
results for 2023–2030 for some developed (e.g., Japan, Russia,
Australia, the USA, the UK, the UAE, Canada, France, New Zealand,
Saudi Arabia, and Turkey) and developing countries (e.g., China,
Pakistan, India, Bangladesh, Kenya, Egypt, and Iran). These ndings add
to a growing body of literature (e.g., Makridis and Han, 2021), high-
lighting the potential of AI-based innovation to promote gender equality
and empower women and girls. These nations use AI with VR/AR to
empower women and girls economically and emotionally. They have
also created AI-powered platforms to monitor hiring and policymaking
for sexism, and forums to discuss improving laws to protect women and
girls (Makridis and Han, 2021). Still, H2b is not supported by other
developed (e.g., Germany) and developing countries (e.g., Nigeria and
Uzbekistan). Therefore, H2b is supported overall, as only three countries
did not support the hypothesis. Moreover, since most developed coun-
tries are anticipated to show better performance in achieving SDG5
compared to developing countries, H4e, which proposed that the better
institutional arrangements in developed countries help them perform
better in achieving AI-based innovation and resultant better SDG5, is
supported, overall. This aligns with the research of Alarc´
on and Cole
(2019), which noted higher gender equality achievement in developed
countries compared to developing countries.
Sixth, H3a, which proposed that AI-based innovation (SDG9) posi-
tively inuences SDG6 (clean water and sanitation for all), is supported
by the estimated results for 2023–2030 for certain developed (e.g.,
Australia, France, Germany, Japan, Russia, Saudi Arabia, and the UAE)
and developing countries (e.g., Egypt, Iran, Nigeria, Pakistan, Vietnam,
20
All developing countries except one are projected to show an increasing
trend in the period considered.
21
Only two developed countries did not support H1c.
22
All developing countries are projected to show an increasing trend in the
period considered.
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
22
Kenya, Bangladesh, China, and India). The ndings of Kumar et al.
(2021) align with this, highlighting these countries’ policies, such as
managing water infrastructure using an Automatic Identication System
(AIS), IoT for water installation monitoring, and big data for water usage
analysis. Yet, H3a is not supported for other developed (e.g., the USA,
Canada, UK, New Zealand, and Turkey) and developing countries (e.g.,
Uzbekistan). Thus, H3a is partially supported. Moreover, since devel-
oping countries are anticipated to show better performance in achieving
SDG6 compared to developed countries, H4f, which proposed that better
institutional arrangements in developed countries help them perform
better in achieving AI-based innovation and resultant better SDG6, is not
supported. This contradicts the research of Fuenfschilling and Binz
(2018), which reported that developed countries with better institu-
tional arrangements typically demonstrate higher technology-based
innovation and consequent better achievement in SDG6.
Seventh, H3b, which proposed that AI-based innovation (SDG9)
positively inuences SDG7 (affordable and clean energy), is supported
by the estimated results for 2023–2030 for some developed (e.g., France,
New Zealand, Saudi Arabia, the UK, Australia, Japan, Germany, Turkey,
and the USA) and developing countries (e.g., Egypt, Uzbekistan, Viet-
nam, Kenya, Bangladesh, China, and India). This is consistent with the
ndings of Zhu et al. (2020), who further state that countries are using
blockchain technology to better monitor the return on investment in
clean & renewable energy, automation to lower energy facility inspec-
tion costs, and digital educational platforms for alternative energy use.
Nevertheless, H3b is not supported by other developed (e.g., Russia,
Canada, and UAE) and developing countries (e.g., Pakistan, Iran, and
Nigeria). Consequently, H3b is partially supported. Moreover, since
both developed and developing countries are anticipated to show similar
performance in achieving SDG7, H4g, which proposed that better
institutional arrangements in developed countries help them perform
better in achieving AI-based innovation and resultant better SDG7, is not
supported. This contradicts the research of Makridis and Han (2021),
which states that better institutional quality, such as improved gover-
nance and regulation, contributes to superior SDG7 achievement in
developed countries.
Eighth, H1d, which proposed that AI-based innovation (SDG9)
positively inuences SDG8 (decent work and economic growth), is
supported by the estimated results for 2023–2030 for some developed
(e.g., France, New Zealand, UAE, Russia, the UK, Germany, and Turkey)
and developing countries (e.g., Uzbekistan, Nigeria, Iran, Egypt, Kenya,
Bangladesh, and China). This is consistent with the ndings of Rodrí-
guez-Espíndola et al. (2022), who further state that these countries
provide free platforms for professionals, independent contractors, and
freelancers, creating new opportunities for decent work and economic
growth. Still, H1d is not supported for other developed (e.g., Saudi
Arabia, Australia, the USA, Japan, and Canada) and developing coun-
tries (e.g., Uzbekistan and Vietnam). Hence, H1d is partially supported.
Moreover, since developing countries are anticipated to show better
performance in achieving SDG8 compared to developed countries, H4h,
which proposed that the better institutional arrangements in developed
countries help them perform better in achieving AI-based innovation
and resultant better SDG8, is not supported. This is consistent with the
ndings of Fuenfschilling and Binz (2018), indicating a complex rela-
tionship between institutional arrangements and SDG8, and that better
institutional arrangements in developed countries do not necessarily
guarantee better performance in achieving decent work and economic
growth.
Ninth, H2c, which proposed that AI-based innovation (SDG9) posi-
tively inuences SDG10 (reduced inequality), is supported by the esti-
mated results for 2023–2030 for some developed (e.g., Japan, France,
Canada, Russia, and the UAE) and developing countries (e.g., Nigeria,
Kenya, and China). This is in line with the ndings of Garg et al. (2022),
who further state that these countries are consistently implementing
measures such as accessible digital platforms for learning, smart terri-
tories to encourage citizens’ participation in urban decisions, and
interconnected Internet banking systems to combat digital and eco-
nomic divides. Nevertheless, H2c is not supported by other developed (e.
g., the USA, the UK, Saudi Arabia, Australia, Germany, New Zealand,
Turkey) and developing countries (e.g., Uzbekistan, Pakistan, Egypt,
Bangladesh, Iran, and India). Thus, H2c is not supported overall, as most
countries do not support the hypothesis. However, since developed
countries are anticipated to show better performance in achieving
SDG10 compared to developing countries, H4i, which proposed that the
better institutional arrangements in developed countries help them
perform better in achieving AI-based innovation and resultant better
SDG10, is supported overall. Zall´
e (2019) supports this, suggesting that
developed countries with superior institutional quality and efcient
governance structures can implement policies and initiatives to reduce
inequality through social welfare programs, progressive tax systems,
and targeted interventions in marginalized communities.
Tenth, H2d, which proposed that AI-based innovation (SDG9) posi-
tively inuences SDG11 (sustainable cities and communities), is sup-
ported by the estimated results for 2023–2030 for some developed (e.g.,
Canada, the USA, Australia, New Zealand, Russia, and the UAE) and
developing countries (e.g., India, Vietnam, Bangladesh, Pakistan, Kenya,
and China). This is consistent with the ndings of Ullah et al. (2021),
which highlight these countries’ implementation of smart cities with
IoT-backed action plans for urban emergencies and the use of urban data
blockchain technologies for recovery contingencies. Nevertheless, H2d
is not supported by other developed (e.g., the UK, Germany, and Turkey)
and developing countries (e.g., Uzbekistan, Nigeria, Japan, Saudi Ara-
bia, Egypt, and Iran). Hence, H2d is partially supported. However, since
developed countries are anticipated to show better performance in
achieving SDG11 compared to developing countries, H4j, which pro-
posed that the better institutional arrangements in developed countries
help them perform better in achieving AI-based innovation and resultant
better SDG11, is supported overall. This is supported by Gupta et al.
(2023), who found that developed countries with strong institutional
arrangements have better urban policies and governance structures,
resulting in better outcomes in sustainable urban development.
Eleventh, H3c, which proposed that AI-based innovation (SDG9)
positively inuences SDG12 (responsible consumption and production),
is supported by the estimated results for 2023–2030 for one developing
country, namely, Vietnam. This is consistent with the ndings of Wang
et al. (2020), who state that countries showing good scores in achieving
SDG12 are implementing advice-rendering information systems for
agricultural sustainability, environmental certication for global market
supply, robotics for ‘virtual’ experiments in manufacturing processes,
big data for manufacturing and consumption patterns, and AI-based
sensing. However, H3c is not supported for most other developed (e.
g., Australia, Canada, Germany, Japan, New Zealand, Saudi Arabia,
France, Turkey, the USA, the UK, the UAE, and Russia) and developing
countries (e.g., Iran, India, Bangladesh, Kenya, Nigeria, Pakistan, Egypt,
China, and Uzbekistan). Consequently, H3c is not supported. This is in
line with the research by Dolezal and Novelli (2022), who found that the
relationship between AI and sustainable development is complex and
context-specic, with mixed results across different countries. However,
since developing countries are anticipated to show better performance
in achieving SDG12 compared to developed countries, H4k, which
proposed that the better institutional arrangements in developed coun-
tries help them perform better in achieving AI-based innovation and
resultant better SDG12, is not supported. This is contradicted by the
ndings of Johansen and Vestvik (2020), who found that developed
countries tend to have better institutional arrangements for promoting
innovation, including nancial support for research and development,
intellectual property protection, and policies that encourage responsible
consumption and production.
Twelfth, H3d, which proposed that AI-based innovation (SDG9)
positively inuences SDG13 (climate action), is supported by the esti-
mated results for 2023–2030 for some developed (e.g., France,
Australia, Germany, Japan, New Zealand, Canada, Turkey, the UAE, the
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
23
UK, the USA, Saudi Arabia, and Russia) and developing countries (e.g.,
Nigeria, Egypt, and Vietnam). As Kumar et al. (2020) note, these
countries utilize digital tools for environmental education, advanced AI
for survival education, smart cities for pollution reduction, and AI with
robotics for environmental monitoring. However, H3d is not supported
by developing countries (e.g., Uzbekistan, China, India, Kenya,
Bangladesh, Iran, and Pakistan). Hence, H3d is partially supported. This
nding is consistent with other studies conducted by Di Vaio et al.
(2020), who also found challenges for developing countries in fully
embracing AI-based innovation for sustainable development. Moreover,
since developed countries are expected to outperform developing
countries in achieving SDG13, H4l, which proposed that the better
institutional arrangements in developed countries help them perform
better in achieving AI-based innovation and resultant better SDG13, is
supported. This is consistent with the ndings of Johansen and Vestvik
(2020), who reported that institutional factors in developed countries,
such as congenial government support, innovation policies, and intel-
lectual property rights, signicantly inuence the successful imple-
mentation of AI-based innovation for climate action-related sustainable
development.
Thirteenth, H3e, which proposed that AI-based innovation (SDG9)
positively inuences SDG14 (Life below water), is supported by the
estimated results for 2023–2030 for some developed (e.g., Germany, the
UK, the USA, and France) and developing countries (e.g., Iran, and
Pakistan). This is in line with the ndings of Wang et al. (2020), who
further state that these countries are using AI-based drones and IoT to
expose predatory shing and environmental damage (e.g., plastic
dumped into the sea). Big data and AIS for research as well as marine
resource protection, were other measures. However, H3e is not sup-
ported for other developed (e.g., Saudi Arabia, Japan, Canada, New
Zealand, Russia, the UAE, and Turkey) and developing countries (e.g.,
Uzbekistan, China, India, Nigeria, Bangladesh, Egypt, and Kenya). Thus,
H3e is not supported, as most countries do not support this hypothesis.
This nding is in line with previous research (e.g., Johansen and Vestvik,
2020) that noted the challenges in achieving SDG14, particularly in
developing countries where there may be limited resources and insti-
tutional arrangements. Moreover, since developed and developing
countries are forecast to show almost similar performance in achieving
SDG14, H4m, which proposed that better institutional arrangements in
developed countries help them achieve better AI-based innovation and
resultant better SDG14, is not supported. Gupta et al. (2023) corroborate
these ndings, stating that developing countries are progressing towards
SDG14 despite the challenges, with technology transfer and capacity
building playing crucial roles in promoting sustainable development.
Fourteenth, H3f, which proposed that AI-based innovation (SDG9)
positively inuences SDG15 (Life on land), is supported by the estimated
results for 2023–2030 for some developed (e.g., Australia, Canada, the
UAE, the UK, France, and the USA) and developing countries (e.g.,
Uzbekistan). This corroborates the ndings by Kazancoglu et al. (2021),
who further state that these nations are using AI to optimize plantation-
related water consumption and IoT-enabled smart sensors for real-time
detection. They use aerial imagery and robotics to detect crop diseases
and early res using machine learning. Nevertheless, H3f is not sup-
ported for other developed (e.g., Japan, Germany, Russia, Turkey, New
Zealand, and Saudi Arabia) and developing countries (e.g., Vietnam,
Egypt, Pakistan, Bangladesh, China, India, Iran, Kenya, and Nigeria).
Consequently, H3f is not supported, as most countries do not support
this hypothesis. However, since developed countries are anticipated to
show better performance in achieving SDG15 than developing countries,
H4n, which proposed that the better institutional arrangements in
developed countries help them perform better in achieving AI-based
innovation and resultant better SDG15, is supported. This aligns with
previous research that indicates that (developed) countries with better
institutional arrangements tend to perform better in achieving the sus-
tainable coexistence of plants and animals on land through smart utili-
zation of AI-based innovation compared to developing countries (Yin
et al., 2022).
Fifteenth, H2e, which proposed that AI-based innovation (SDG9)
positively inuences SDG16 (peace, justice and strong institutions), is
supported by the estimated results for 2023–2030 for some developed
(e.g., France, Saudi Arabia, the UAE, Japan, Russia, and Turkey) and
developing countries (e.g., Iran, China, Egypt, Pakistan, Vietnam, and
Uzbekistan). This is consistent with the ndings of Klofsten et al. (2019),
who further state that these nations are implementing digital educa-
tional platforms and integrated digital criminal evidence systems, and
using AI in conjunction with drones, IoT, and blockchain to create in-
tegrated crime-ghting tools and combat institutional corruption.
Nevertheless, H2e is not supported by other developed (e.g., Australia,
the UK, Canada, the USA, Germany, and New Zealand) and developing
countries (e.g., India, Bangladesh, Kenya, and Nigeria). Therefore, H2e
is partially supported. However, since developing countries are antici-
pated to show better performance in achieving SDG16 than developed
countries, H4o, which proposed that the better institutional arrange-
ments in developed countries help them perform better in achieving AI-
based innovation and resultant better SDG16, is not supported. This is
supported by some recent literature (e.g., Chimhowu et al., 2019)
nding that developing countries may outperform developed countries
in achieving SDG16, despite having comparatively weaker institutional
arrangements.
Sixteenth, H2f, which proposed that AI-based innovation (SDG9)
positively inuences SDG17 (partnerships for the goals), is supported by
the estimated results for 2023–2030 for some developed (e.g., New
Zealand, Australia, Canada, France, Germany, Japan, and Turkey) and
developing countries (e.g., India, Iran, China, Nigeria, Pakistan, Viet-
nam, and Uzbekistan). This parallels the study by Klofsten et al. (2019),
who state that these nations use virtual platforms for planning innova-
tion, VR/AR for educational and international alliances, and AI with big
data for national development. Nevertheless, H2f is not supported by
developed (e.g., Russia, the USA, Saudi Arabia, the UK, and the UAE)
and developing countries (e.g., Egypt, Bangladesh, and Kenya). Hence,
H2f is partially supported. Nevertheless, since developing countries are
anticipated to show better performance in achieving SDG17 than
developed countries, H4p, which proposed that the better institutional
arrangements in developed countries help them perform better in
achieving AI-based innovation and resultant better SDG17, is not sup-
ported. This is supported by the ndings of Chimhowu et al. (2019), who
state that AI-based innovation can help to identify gaps in existing
partnerships, highlight areas for potential collaboration, and facilitate
the sharing of information and resources among stakeholders.
5.2. Contributions and implications to research
5.2.1. Theoretical contributions
This paper makes several theoretical contributions by addressing the
research gap articulated in the introduction. First, this study used
Fountain’s (2001) TEF model to examine how institutional, organiza-
tional, technical, and situational variables affect SDG attainment across
countries. As part of that model, this paper demonstrates the impact of
particular institutional arrangements and organizational structures on
the consequent technology’s characteristics (enacted technologies). The
paper demonstrates the dynamic relationships between implemented
technology (innovation), organizational forms (AI adoption), and insti-
tutional arrangements (country-specic contextual factors). In the cur-
rent model (theory), technology enactments (innovation) are created
through a dynamic process in which organizational characteristics (AI
adoption) or the presence of valid formal procedures (institutional ar-
rangements) may either facilitate or impede a given enactment (char-
acteristics of the technology). Along these lines, the study’s key ndings
appear to be consistent with what many consider to be the need to
integrate innovation with essential complementary resources (such as
AI) to increase the value of these technologies in achieving sustainability
(Bag et al., 2021). Hence, this study’s key theoretical contribution is to
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
24
extend TEF to demonstrate the coexistence of AI and creativity in
accomplishing SDGs. As a result, this study’s most signicant theoretical
contribution is to present an extension of TEF that illustrates the co-
existence of AI and innovation in accomplishing SDGs. Secondly,
although the roles played by these constructs may vary depending on the
context and each of these constructs has been studied separately in the
past, studying their interaction is novel and makes a substantial
contribution to the existing literature on AI, Innovation, Sustainable
Development, System Dynamics, SDGs, and the TEF literature. It is
worth mentioning that examining any of these constructs independently
may result in a limited comprehension of the SDG achievement process
as a whole, as per prior research (Luna-Reyes and Gil-Garcia, 2011;
Malodia et al., 2021). Thirdly, this study also adds to the existing
literature on AI, Innovation, and SDGs by fostering fresh viewpoints on
the impact of AI-based innovation on SDGs at different levels. At the 1st
level, the direct impact of innovation on SDGs was investigated. At the
2nd level, the “enabling effects” of AI on innovation were investigated in
its causal impact on SDGs. At the 3rd level, the long-term impact of AI-
based innovation on SDGs was estimated. Also, this study makes an
original contribution to the SDG and sustainability literature by exam-
ining the intricate relationship between SDG9 (innovation-related SDG)
and the other SDGs. Fourthly, the paper proposes a comprehensive
model that integrates the effects of AI-based innovation on SDGs using a
system dynamics perspective. This model takes into account the com-
plex interconnections and feedback loops between various aspects of AI-
based innovation and SDGs, which have not previously been explored in
a systematic way. Fifthly, this study also enhances the literature on
system dynamics by exploring how Systems Dynamics can be employed
for forecasting complex problems of interaction between AI, innovation,
and SDGs for around a decade for 22 different nations covering ve
different continents. In fact, this paper combines institutional theory
(Technology Enactment Framework) and system dynamics to explain
complex SDG achievement. Very few previous studies have utilized this
methodology. This is another contribution of this paper. Sixthly, this
whole analysis has been done from a comparative perspective between
developed and developing countries. Therefore, it contributes to the
literature relating to developed and developing countries. Finally, this
research combines institutional theory (technology enactment frame-
work) with simulations based on system dynamics to better understand
SDG phenomena. Institutional theory is strong, but its theoretical
abstraction has questioned its application. Institutional theory and sys-
tem dynamics may help researchers precisely dene variables and hy-
pothesized linkages. Mathematical accuracy in computer simulation
requires variables to be conceptually and theoretically operationalized
to develop a system dynamics model. Thus, institutional theory’s
complexity and critical notions are explored, while the approach’s ca-
pacity to operationalize and interpret them concretely is retained. As
previously stated, the simulation model may be regarded as a theory and
assessed for internal consistency.
5.2.2. Practical implications
The paper makes several practical contributions. First, although the
SDGs (which represent an economic opportunity worth 12 trillion dol-
lars) promise improved nancial performance, it can be complicated to
make the goals tangible at the business level because of the sheer
magnitude of the SDGs (United Nations, 2022a). If used wisely, AI-based
innovation can provide businesses with the foundation for the prioriti-
zation of SDGs and enable them to translate global goals down to an
organizational level (Lee et al., 2022). Second, this study elucidates the
need for collaboration among academia, industry, and government for
SDG-supporting technology advances to meet the SDG 2030 objective. I
emphasize the importance of these players making more investments in
AI-driven innovation and giving access to high-quality data for this
purpose. Cleaner manufacturing is particularly vital for sustainable
economic and environmental reasons. Third, the study highlights the
need for continued investment in AI research and development to spur
innovation for the SDGs. Accordingly, governments should allocate re-
sources for developing AI technologies relevant to their specic SDG-
related challenges. Fourth, Governments can invest in capacity-
building programs to develop the skills needed to harness the poten-
tial of AI for the SDGs. This will ensure they have the expertise to
develop and implement AI-driven solutions effectively. Fifth, Govern-
ments can use the model presented in the study to monitor and evaluate
the impact of AI-driven innovations on the SDGs. This will help to
identify areas that require improvement and guide the development of
more effective measures. Sixth, the model uses a cross-country setting,
indicating that the impact of AI-driven innovations on SDGs is not
limited to one country. Decision-makers should use this insight to
compare their country’s progress against other countries and identify
priority areas for improvement. Governments should also collaborate
with other countries to share best practices, experiences, and in-
novations in AI-driven sustainable development. Seventh, the study
emphasizes the importance of data-driven decision-making in achieving
the SDGs. Governments can use the model’s predictions and recom-
mendations to make informed decisions backed by data and evidence.
5.2.3. Policy implications
The paper makes several policy contributions. First, the study dem-
onstrates the potential of AI-based innovations in achieving the SDGs.
Governments can use this information to integrate AI-driven solutions
into national development plans. This can help to identify the need for
resources for implementing AI-driven innovations that will signicantly
impact on the SDGs. Second, the paper’s model provides decision-
makers with a fact-based assessment of the impact of AI-based innova-
tion on SDGs. This can help associated governments in 22 countries
establish efcient long-term policies and measures for sustainable
development in the economic, social, and environmental domains. This
can help to identify potential challenges and opportunities for different
countries and develop targeted policies and interventions addressing
their needs. Third, policymakers can use the model to identify and pri-
oritize the most promising AI-driven innovations based on their poten-
tial to contribute to the SDGs. Fourth, the paper can help policymakers
and entrepreneurs interested in creating new markets for SDG-related
products, such as electricity, agriculture, legal, social, and urban ser-
vices. The study provides insights into the potential impact of AI-based
innovation on these sectors and can help experts prioritize their in-
vestments in these areas. Fifth, this study can also be used as a reference
by policymakers to help them plan their new AI investments more
effectively and spur more innovation for the SDGs. Sixth, the ndings of
this study can help policymakers and entrepreneurs identify new op-
portunities for collaboration in terms of AI-based innovation in the eld
of sustainable development in various developed and developing
nations.
5.2.4. Social implications
Since AI-related concerns are both complex and multidimensional,
the ndings of this study are consistent with the idea and context of ‘AI
for Good.’ Therefore, the study’s predictions suggest that we should use
AI to improve people’s lives and reduce inequalities. All sustainable
development initiatives should have human survival and growth as their
end goal. Using the SDGs as a framework for understanding sustainable
development, it has been demonstrated that linking AI-based innovation
to the SDGs can shed light on solving the signicant challenges faced by
AI, such as ensuring the safety and dignity of individuals. In terms of
obvious global concerns, the World Economic Forum has identied
cybersecurity as a top concern (World Economic Forum, 2020). As a
result of this research, policymakers in newly industrialized and dem-
ocratic societies in developing nations may focus on transparency,
ethics, and full citizen involvement as cornerstones of effective AI
governance. The SDGs are based on universal principles like diversity,
inclusion, and ‘leaving no one behind,’ and my ndings align with those
principles.
S. Nahar
Technological Forecasting & Social Change 201 (2024) 123203
25
5.3. Limitations and avenues for further research
The study has certain limitations, which suggest avenues for future
research. First, because the results are predicted using a simulation
model, the SD method used for prediction has some aws. This study’s
ndings should, therefore, be interpreted in light of prior research on the
topic. Future studies should also use empirical research to obtain factual
results. Second, due to its 22-country sample size, this research has some
limitations. Thus, to better understand AI-driven innovations, I recom-
mend replicating them in other countries, especially those with less SDG
effort, in future studies. Third, the study is conducted across different
countries with varying levels of development and political systems.
Hence, it may be difcult to generalize the results to other countries or
regions. Future research can concentrate on specic nations and eval-
uate the effect of AI-based innovation on achieving SDGs in the local
context in greater detail. Fourth, the study only considers the impact of
AI-based innovation on SDGs at the macro-level, and future research
should consider exploring the impact at the micro-level, such as the
effects on individual households and communities. Fifth, the results of
using AI-driven innovation to implement the SDG 2030 Agenda may
seem too ambitious without input from other sources, like qualitative
research with developers and experts of AI-driven innovation or focus
groups where policymakers talk about these results. Future studies
should capture the aforementioned qualitative insights. Sixth, I have not
considered organizational forms, institutional arrangements, enacted
technologies and outcomes’ recursive relationships. Future studies
exploring the interaction between AI, innovation, and SDGs should
explore these recursive relationships. In other words, future research
should consider exploring how SDGs can drive AI-based innovation and
what the potential barriers are to this relationship. Seventh, future
research should consider the interdependence of SDGs and how the
impact of AI-based innovation on one SDG can affect others.
CRediT authorship contribution statement
Sharmin Nahar: Conceptualization, Data curation, Formal analysis,
Methodology, Project administration, Resources, Validation, Visualiza-
tion, Writing – original draft, Writing – review & editing.
Declaration of competing interest
I conrm that I do not have any ties or engagement with any orga-
nization or entity that has a nancial or non-nancial interest in the
topics or materials explored in this manuscript. Additionally, no orga-
nizations have provided assistance, nancial or otherwise, for the work
submitted.
Data availability
Data will be made available on request.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.techfore.2023.123203.
References
Abbasi, K.R., Hussain, K., Haddad, A.M., Salman, A., Ozturk, I., 2022. The role of
nancial development and technological innovation towards sustainable
development in Pakistan: fresh insights from consumption and territory-based
emissions. Technol. Forecast. Soc. Chang. 176, 121444 https://doi.org/10.1016/j.
techfore.2021.121444.
Akizu-Gardoki, O., Bueno, G., Wiedmann, T., Lopez-Guede, J.M., Arto, I., Hernandez, P.,
Moran, D., 2018. Decoupling between human development and energy consumption
within footprint accounts. J. Clean. Prod. 202, 1145–1157. https://doi.org/
10.1016/j.jclepro.2018.08.235.
Alarc´
on, D.M., Cole, S., 2019. No sustainability for tourism without gender equality.
J. Sustain. Tour. 27 (7), 903–919. https://doi.org/10.1080/
09669582.2019.1588283.
Amabile, T.M., 2020. Creativity, articial intelligence, and a world of surprises. Acad.
Manag. Discov. 6 (3), 351–354. https://doi.org/10.5465/amd.2019.0075.
Aparicio, S., Urbano, D., Stenholm, P., 2021. Attracting the entrepreneurial potential: a
multilevel institutional approach. Technol. Forecast. Soc. Chang. 168, 120748
https://doi.org/10.1016/j.techfore.2021.120748.
Bag, S., Pretorius, J.H.C., Gupta, S., Dwivedi, Y.K., 2021. Role of institutional pressures
and resources in the adoption of big data analytics powered articial intelligence,
sustainable manufacturing practices and circular economy capabilities. Technol.
Forecast. Soc. Chang. 163, 120420 https://doi.org/10.1016/j.
techfore.2020.120420.
Barba-S´
anchez, V., Arias-Antúnez, E., Orozco-Barbosa, L., 2019. Smart cities as a source
for entrepreneurial opportunities: evidence for Spain. Technol. Forecast. Soc. Chang.
148, 119713 https://doi.org/10.1016/j.techfore.2019.119713.
Benzidia, S., Makaoui, N., Bentahar, O., 2021. The impact of big data analytics and
articial intelligence on green supply chain process integration and hospital
environmental performance. Technol. Forecast. Soc. Chang. 165, 120557 https://
doi.org/10.1016/j.techfore.2020.120557.
Berente, N., Gu, B., Recker, J., Santhanam, R., 2021. Managing artificial intelligence. MIS
Q. 45 (3), 1433–1450.
Bhimani, A., Hausken, K., Arif, S., 2022. Do national development factors affect
cryptocurrency adoption? Technol. Forecast. Soc. Chang. 181, 121739 https://doi.
org/10.1016/j.techfore.2022.121739.
Black, L.J., 2013. When visuals are boundary objects in system dynamics work. Syst.
Dyn. Rev. 29 (2), 70–86. https://doi.org/10.1002/sdr.1496.
Burchardt, H.-J., Ickler, J., 2021. Time to live well: well-being and time afuence for
sustainable development. Third World Q. 42 (12), 2939–2955. https://doi.org/
10.1080/01436597.2021.1981761.
Calabrese, A., Costa, R., Tiburzi, L., Brem, A., 2023. Merging two revolutions: a human-
articial intelligence method to study how sustainability and industry 4.0 are
intertwined. Technol. Forecast. Soc. Chang. 188, 122265 https://doi.org/10.1016/j.
techfore.2022.122265.
Cernev, T., Fenner, R., 2020. The importance of achieving foundational sustainable
development goals in reducing global risk. Futures 115, 102492. https://doi.org/
10.1016/j.futures.2019.102492.
Chimhowu, A.O., Hulme, D., Munro, L.T., 2019. The ‘new’ national development
planning and global development goals: processes and partnerships. World Dev. 120,
76–89. https://doi.org/10.1016/j.worlddev.2019.03.013.
Cockburn, I.M., Henderson, R., Stern, S., 2019. 4. In: The Impact of Articial Intelligence
on Innovation: An Exploratory Analysis. In 4. The Impact of Articial Intelligence on
Innovation: An Exploratory Analysis. University of Chicago Press, pp. 115–148.
https://doi.org/10.7208/9780226613475-006.
Cosenz, F., Rodrigues, V.P., Rosati, F., 2020. Dynamic business modeling for
sustainability: exploring a system dynamics perspective to develop sustainable
business models. Bus. Strateg. Environ. 29 (2), 651–664. https://doi.org/10.1002/
bse.2395.
de Sousa Jabbour, A.B.L., Ndubisi, N.O., Seles, Roman Pais, B. M., 2020. Sustainable
development in Asian manufacturing SMEs: Progress and directions. Int. J. Prod.
Econ. 225, 107567 https://doi.org/10.1016/j.ijpe.2019.107567.
Del Vecchio, P., Secundo, G., Maruccia, Y., Passiante, G., 2019. A system dynamic
approach for the smart mobility of people: implications in the age of big data.
Technol. Forecast. Soc. Chang. 149, 119771 https://doi.org/10.1016/j.
techfore.2019.119771.
Demiralay, S., Gencer, H.G., Bayraci, S., 2021. How do articial intelligence and robotics
stocks co-move with traditional and alternative assets in the age of the 4th industrial
revolution? Implications and Insights for the COVID-19 period. Technol. Forecast.
Soc. Chang. 171, 120989 https://doi.org/10.1016/j.techfore.2021.120989.
Dhahri, S., Slimani, S., Omri, A., 2021. Behavioral entrepreneurship for achieving the
sustainable development goals. Technol. Forecast. Soc. Chang. 165, 120561 https://
doi.org/10.1016/j.techfore.2020.120561.
Di Vaio, A., Palladino, R., Hassan, R., Escobar, O., 2020. Articial intelligence and
business models in the sustainable development goals perspective: a systematic
literature review. J. Bus. Res. 121, 283–314. https://doi.org/10.1016/j.
jbusres.2020.08.019.
Dolezal, C., Novelli, M., 2022. Power in community-based tourism: empowerment and
partnership in Bali. J. Sustain. Tour. 30 (10), 2352–2370. https://doi.org/10.1080/
09669582.2020.1838527.
Elsayed, M., Ran, Y., Ai, P., Azab, M., Mansour, A., Jin, K., Zhang, Y., Abomohra, A.E.-F.,
2020. Innovative integrated approach of biofuel production from agricultural wastes
by anaerobic digestion and black soldier y larvae. J. Clean. Prod. 263, 121495
https://doi.org/10.1016/j.jclepro.2020.121495.
Forrester, J.W., 1958. Industrial dynamics: a major breakthrough for decision makers.
Harv. Bus. Rev. 36 (4), 37–66.
Forrester, J.W., Mass, N.J., Ryan, C.J., 1976. The system dynamics national model:
understanding socio-economic behavior and policy alternatives. Technol. Forecast.
Soc. Chang. 9 (1), 51–68. https://doi.org/10.1016/0040-1625(76)90044-5.
Fountain, J.E., 1995. Enacting Technology: An Institutional Perspective. John F.
Kennedy School of Government, Harvard University, MA.
Fountain, J.E., 2001. Building the Virtual State: Information Technology and
Institutional Change. Brookings Institution Press.
Fritz, M.M., Ruel, S., Kallmuenzer, A., Harms, R., 2021. Sustainability management in
supply chains: the role of familiness. Technol. Forecast. Soc. Chang. 173, 121078
https://doi.org/10.1016/j.techfore.2021.121078.
S. Nahar