AI Leads, Cybersecurity Follows: Unveiling Research Priorities in Sustainable Development Goal-Relevant Technologies across Nations

Abstract

This study presents a global analysis of research priorities for technologies relevant to Sustainable Development Goals (SDGs). We examine 18 technological domains across countries, introducing a novel within-country rank metric to normalize differences in research output. Using a combination of linear regression and K-means cluster analysis, we identify factors influencing overall productivity and reveal distinct patterns in research priorities among nations. Our analysis of Web of Science total publication data yields five country clusters with specific technological focus areas: Eco-Tech Innovators, Cyber-Digital Architects, Bio-Industrial Pioneers, Geo-Data Security Analysts, and Cyber-Sustainable Integrators. We find that while economic indicators strongly predict overall research productivity, countries with similar economic profiles often exhibit divergent research priorities. Artificial Intelligence emerges as a top priority across all clusters, while areas such as blockchain and digital twins show lower prioritization despite their theoretical importance. Our findings reveal unexpected similarities in research focus among geopolitically diverse countries and highlight regional patterns in technological emphasis. This study offers valuable information for policymakers and researchers, enhancing our understanding of the global landscape of SDG-relevant technological research and potential avenues for international collaboration.

1. Introduction

The United Nations Sustainable Development Goals (SDGs) aim to address global challenges such as poverty, inequality, climate change, and environmental degradation. Technology plays a key role in achieving these goals by offering new solutions to development issues and creating opportunities for sustainable growth. Various technologies, from Artificial Intelligence and Big Data Analytics to renewable energy and biotechnology, can improve resource use, healthcare, education, and sustainable production. These technologies also help in tracking progress towards the SDGs by providing data for decision-making (UNDP, 2023) [1].

Given the importance of technology for the SDGs, it is necessary to understand how different countries contribute to and prioritize research in these areas. A country’s research output and focus on SDG-relevant technologies show its commitment to sustainable development and its ability to create innovative solutions to global problems. This study analyzes research productivity and priorities in SDG-relevant technologies across countries. We examine publication data in 18 technological domains important for sustainable development to identify patterns in research focus among nations. We also relate this research output to socio-economic indicators and human development factors to understand what drives technological innovation for sustainable development.

This analysis is important for several reasons. First, it identifies patterns in research priorities across countries, revealing how nations focus their efforts on different SDG-relevant technologies. This information can help in recognizing potential areas for international collaboration and knowledge sharing. Second, by examining the relationship between research output and multiple development indicators, we can better understand the factors that influence a country’s capacity for innovation in sustainable technologies. This knowledge is valuable for policymakers and funding agencies in shaping research strategies. Third, our approach combines broad correlations identified through regression analysis with typological and regional insights revealed by cluster analysis. This examination allows us to go beyond simple productivity metrics and uncover unexpected research specializations, particularly in developing countries. Lastly, by mapping the global landscape of SDG-related technological research, we can identify potential gaps and areas needing increased attention or investment. As we work to achieve the SDGs by 2030, understanding the current state of technological research is essential for using innovation to create a more sustainable and fair world.

One of the primary challenges in achieving the SDGs is the persistent digital divide, which exacerbates existing inequalities (ITU and UNDP, 2023) [2]. Technologies can help bridge this gap by expanding affordable and reliable connectivity to underserved areas, ensuring universal access to digital services (United Nations, 2023) [3]. Additionally, technologies can address data challenges by improving the volume, quality, and interoperability of data, particularly in low-income geographies and among vulnerable populations (World Economic Forum, 2020) [4]. Another significant challenge is the lack of adequate finance infrastructure to support digital and decentralized technologies (World Economic Forum, 2020) [4]. Technologies can contribute to overcoming this by enabling innovative financial models and improving access to financial services for marginalized communities (ITU and UNDP, 2023) [2].

Furthermore, technologies can help address governance and policy challenges by enhancing transparency, accountability, and efficiency in public institutions (United Nations, 2023) [3]. They can also support efforts to combat climate change and environmental degradation by enabling more sustainable practices across industries (World Economic Forum, 2020) [4].

According to the reports, several specific technologies are particularly relevant for overcoming these challenges. Artificial Intelligence (AI) and machine learning can enhance decision-making processes, improve resource allocation, and provide personalized services in areas such as healthcare and education (ITU and UNDP, 2023) [2]. AI can also contribute to climate action by optimizing energy consumption and supporting environmental monitoring (World Economic Forum, 2020) [4]. The Internet of Things (IoT) enables the implementation of smart systems for urban management, agriculture, and resource conservation. IoT devices can collect real-time data to inform policy decisions and improve service delivery (United Nations, 2023) [3]. Big Data Analytics supports evidence-based policymaking by processing large volumes of data to identify patterns and trends. This technology is crucial for monitoring progress towards the SDGs and adapting strategies accordingly (World Economic Forum, 2020) [4].

Technological advancements offer significant potential for achieving Sustainable Development Goals (SDGs), as highlighted by various reports. The ITU and UNDP (2023) [2] emphasize the importance of robust cybersecurity measures in enhancing public trust and protecting critical infrastructure. Renewable energy technologies are crucial for addressing climate change and promoting sustainable energy access (ITU and UNDP, 2023) [2]. Geographic Information Systems (GISs) support urban planning and environmental conservation efforts (UNDP, 2023) [1], while blockchain technology enhances transparency in supply chains and governance (ITU and UNDP, 2023) [2]. Next-generation networks, including 5G, improve connectivity and access to digital services (United Nations, 2023) [3]. Biotechnology and genetics contribute to sustainable agriculture and healthcare advancements (World Economic Forum, 2020) [4], and drones support applications like precision agriculture and disaster response (ITU and UNDP, 2023) [2]. Digital twins optimize resource use and improve decision-making in complex systems (World Economic Forum, 2020) [4].

However, these technological interventions also present significant challenges and risks. The World Economic Forum’s 2024 [5] key concerns include the concentration of technological power, which may lead to unequal access to resources, and the increasing potential for AI to escalate conflicts, particularly if integrated into critical decision-making processes. Additionally, the rapid development of AI, including generative models, raises challenges for regulation, as its speed outpaces the creation of safeguards. Other risks include misinformation, job displacement, and the misuse of AI in cyber attacks and weaponry. The World Economic Forum (2020) [4] also notes issues with data quality and interoperability, while the ITU and UNDP (2023) [2] highlight persistent digital divides. Financial barriers and inadequate regulatory frameworks pose obstacles to implementation (United Nations, 2023; World Economic Forum, 2020) [3,4]. Cybersecurity threats are increasing, particularly in developing countries with limited resources (ITU and UNDP, 2023) [2]. The UN (2023) [3] report warns of risks to human rights from state surveillance and predatory business models. Economic inequality may be exacerbated by concentrated investments in technology (United Nations, 2023) [3], and some technologies have high energy demands that could increase greenhouse gas emissions (World Economic Forum, 2020) [4]. The World Economic Forum (2020) [4] also highlights the risk of perpetuating biases in technology design, while the ITU and UNDP (2023) [2] acknowledge the potential for labor market disruption due to automation and AI. These challenges underscore the need for careful planning and governance to ensure technological interventions effectively contribute to SDG achievement without causing unintended harm.

1.1. State of the Art

Previous scientific research also shows that digital technologies play a crucial role in advancing Sustainable Development Goals (SDGs). Among these, AI stands out for its diverse applications, such as enhancing healthcare delivery, optimizing urban planning, and improving decision-making through effective data management (Adel and Alani, 2024; Bachman et al., 2022; Meitei et al., 2023) [6,7,8], and as a general tool for innovation (Obreja, Rughiniș, and Rosner 2024) [9]. Additionally, AI is instrumental in poverty mapping, crop optimization, and predicting climate hazards (Bachman et al., 2022) [7]. Artificial Intelligence (AI) can also positively impact SDGs in areas like education, and infrastructure development (Ametepey et al., 2024; Lampropoulos et al., 2024) [10,11]. AI’s ability to process vast amounts of data and provide insights can aid decision-making and resource optimization across various sustainability challenges (Tripathi et al., 2024) [12]. However, the relationship between AI and sustainable development is complicated. While AI offers opportunities for progress, it also presents risks and challenges that need careful consideration. These include potential negative impacts on certain SDGs, issues of transparency, autonomy, democracy, and privacy (De Falco, 2024; Sætra, 2024) [13,14]. The concentration of AI development in a handful of nations and companies raises concerns about equitable access and potential exacerbation of global inequalities (Sætra, 2024) [14]. Furthermore, as our reliance on AI and interconnected systems grows, cybersecurity becomes a critical concern. The increasing digitalization of critical infrastructure and vital facilities creates vulnerabilities that could be exploited, potentially threatening the very systems designed to support sustainable development (Aljohani, 2024) [15]. To address these challenges, researchers advocate for a more comprehensive approach to assessing AI’s sustainability, considering both its social and environmental impacts and costs (Heilinger et al., 2024) [16]. This approach calls for careful consideration of the ethical implications of AI deployment and emphasizes the need for responsible development and use of AI technologies in pursuit of the SDGs.

Robotics and automation technologies contribute significantly to healthcare by performing tasks like delivering supplies and medications, thereby reducing human involvement in hazardous situations (Kasinathan et al., 2022) [17]. These technologies also promote sustainable production methods and increase efficiency in agriculture and food processing (Hassoun et al., 2022; Vărzaru, 2024) [18,19].

Renewable energy technologies, particularly when integrated with AI and the Internet of Things (IoT), optimize energy production from sustainable sources and help reduce greenhouse gas emissions (Bachman et al., 2022) [7]. Moreover, they support the development of smart grid systems that enable efficient energy management (Bachman et al., 2022) [7]. The IoT itself is essential for real-time data collection, aiding in resource management, urban planning, and healthcare innovations (Adel and Alani, 2024) [6]. Through the collection of biotic and abiotic data, IoT devices also bolster environmental SDGs (Wu et al., 2018) [20].

Geographic Information Systems (GISs) and geospatial technologies provide critical spatial data for agriculture, health monitoring, and environmental management, contributing to several SDGs (Bachman et al., 2022) [7]. Big Data Analytics further supports these goals by enhancing disaster management, improving supply chain traceability, and refining decision-making processes across various sectors (Kasinathan et al., 2022; Hassoun et al., 2022) [17,18].

Cybersecurity plays an important role in pursuing SDGs through digital technologies. It ensures the integrity, confidentiality, and availability of data, which is essential for building trust in digital systems and promoting their widespread adoption for sustainable development efforts (Palomares et al., 2021) [21]. Robust cybersecurity measures protect critical infrastructures, enabling the safe deployment of technologies like AI, IoT, and Big Data Analytics in sectors such as healthcare, energy, and finance (Kasinathan et al., 2022; Michael et al., 2019) [17,22]. By enabling other Industry 4.0 technologies (Hoosain et al., 2020) [23], cybersecurity supports the development of more resilient and sustainable digital ecosystems (Mabkhot et al., 2021) [24]. Effective cybersecurity also helps ensure that countries can safely leverage digital technologies for their sustainable development initiatives.

Biotechnology and genetics research also play a role in health- and agriculture-related SDGs, although specific details on their contributions are not extensively covered in the reviewed literature.

Emerging technologies like Augmented and Virtual Reality (AR/VR) enhance learning, task performance, and information delivery, especially in industrial settings (Mabkhot et al., 2021) [24]. Similarly, 3D printing, or additive manufacturing, allows for customized production and innovative product design, supporting sustainable manufacturing practices (Mabkhot et al., 2021) [24]. The deployment of next-generation networks, particularly 5G, is crucial for the connection of numerous devices with low latency, which is essential for the functioning of smart cities and IoT applications (Palomares et al., 2021) [21].

Biometric identity systems, such as India’s Aadhaar, improve access to services for vulnerable populations (Michael et al., 2019) [22]. Cloud computing, meanwhile, facilitates remote work and collaboration and provides scalable computing resources for various SDG-supporting applications (Kasinathan et al., 2022; Mabkhot et al., 2021; Alexandrescu, 2019) [17,24,25]. Drones and Unmanned Aerial Vehicles (UAVs) are increasingly used for surveillance and delivery in disaster-stricken areas, enhancing response times and minimizing human risk (Kasinathan et al., 2022) [17].

Natural Language Processing (NLP) works in tandem with AI to analyze scientific publications and patents relevant to SDGs, furthering research and innovation (Hajikhani and Suominen, 2022) [26]. Telemedicine and Healthcare technologies, enhanced by AI and IoT, have significantly improved the quality and accessibility of healthcare services (Popkova et al., 2022) [27]. The IoT is a valuable resource for decentralized, evidence-based services in multiple fields (Culic and Radovici, 2017) [28]. Blockchain technology also plays a vital role in ensuring transparency and security in transactions in private and public infrastructures (Stan, Barac and Rosner, 2021; Alexandrescu and Butincu, 2023; Buțincu and Alexandrescu, 2023) [29,30,31], beyond its role in cryptocurrencies (Obreja, 2024) [32], supporting sustainable supply chains, and enabling energy efficiency through peer-to-peer energy trading (Parmentola et al., 2022) [33].

Furthermore, digital twins, which provide real-time digital representations of physical objects, allow for precise monitoring and decision-making across various sectors (Palomares et al., 2021) [21]. Simulation technologies, including product and process simulation, support optimization in manufacturing and enterprise management (Mabkhot et al., 2021) [24].

1.2. Previous Research on Challenges of SDG-Relevant Digital Technologies

The reviewed scientific articles reveal several significant challenges and risks associated with using digital technologies to pursue Sustainable Development Goals. A central concern is cybersecurity. The rapid pace of technological advancement often outpaces the development of adequate cybersecurity measures, creating vulnerabilities in SDG-related initiatives. The interconnected nature of smart city and IoT applications increases the attack surface for cyber threats, potentially compromising large-scale sustainable development projects (Adel and Alani, 2024; Vochescu, Culic and Radovici, 2020) [6,34]. This is critical in sectors like healthcare and energy, where breaches could have devastating consequences (Adel and Alani, 2024; Kasinathan et al., 2022; Palomares et al., 2021) [6,17,21]. Furthermore, the difficulty of protecting critical infrastructures, as discussed by Wu et al. (2018) [20], emphasizes the need for updated cybersecurity strategies. Moreover, the need to balance data protection with the open data sharing required for many SDG initiatives presents a real challenge. Addressing these cybersecurity risks requires continuous updating of security protocols, significant investment in cybersecurity infrastructure and skills, and careful consideration of privacy and ethical concerns in the design and implementation of SDG-focused digital technologies (Michael et al., 2019) [22].

Economic and social disruptions also emerge as relevant challenges. While automation and AI-driven technologies can enhance efficiency, they also risk displacing jobs and amplifying social polarization (Rughiniș, Rughiniș, and Bran, 2024) [35]. This displacement can exacerbate inequality, especially in developing regions where alternative employment opportunities are scarce (di Vaio et al., 2020) [36]. The digital divide further compounds this issue, as disparities in access to advanced technologies limit the ability of less-developed regions to benefit from technological advancements (Popkova et al., 2022; Vyas-Doorgapersad, 2022) [27,37] and may lead to divergent processes of making sense of conflictual situations (Rughiniș and Flaherty, 2022) [38]. Moreover, the technological gradient between large- and small-scale production, highlighted by Mayer-Foulkes, Serván-Mori, and Nigenda (2021) [39], underscores the uneven distribution of technological benefits.

The high costs and resource demands associated with implementing these technologies present additional barriers. Developing and deploying technologies like blockchain and AI require substantial financial investments, which can be prohibitive for smaller enterprises and developing countries (Adel and Alani, 2024; Parmentola et al., 2022) [6,33].

Ethical and regulatory challenges further complicate the deployment of these technologies. The collection and use of large datasets by AI and IoT technologies raise privacy concerns and issues related to the ethical use of AI (Hajikhani and Suominen, 2022; Goh and Vinuesa, 2021) [26,40]. Ensuring these technologies are used responsibly and transparently is very important, yet challenging, as regulations often struggle to keep pace with technological advancements. The rapid pace of innovation frequently outstrips existing regulatory frameworks, making it difficult to align technologies like AI and blockchain with SDG goals effectively (Palomares et al., 2021) [21]. The need for a careful approach to technological adoption, including considerations of ethical implications, is also emphasized by Michael et al. (2019) [22].

Technological complexity and integration issues also pose significant hurdles. For instance, the integration of blockchain with IoT devices presents challenges due to the non-homogeneous nature of network structures and varying computing capacities (De Villiers, Kuruppu, and Dissanayake, 2021) [41]. Additionally, the interdependencies between different technologies, which can have both direct and indirect impacts on various SDGs, require careful management (Mabkhot et al., 2021) [24]. The difficulty of integrating diverse data sources, as discussed by Wu et al. (2018) [20], further complicates the effective deployment of these technologies (Walsh et al., 2021) [42].

Some technologies, despite their potential to contribute to SDGs, also pose environmental risks. The energy-intensive nature of blockchain technologies and the environmental footprint of large data centers required for AI processing can have adverse environmental impacts (Parmentola et al., 2022; Adams, Kewell, and Parry, 2018) [33,43]. The environmental effects of technology deployment, particularly in the context of sustainable food production, are also discussed by Hassoun et al. (2022) [18]. These challenges underscore the need for a careful and balanced approach to deploying digital technologies in pursuit of SDGs, focusing on mitigating these risks to ensure positive outcomes.

Last but not least, scientific research and development are not immune to challenges either. The articles highlight the need for more comprehensive and interdisciplinary research to address gaps in understanding the impacts of technologies on SDGs. Issues such as data quality, availability, and biases present significant barriers to effective technology deployment (Bachman et al., 2022; Hajikhani and Suominen, 2022; ElMassah and Mohieldin, 2020) [7,26,44]. Furthermore, analyzing the impacts of technological innovation is complicated by the interactions between variables such as economic growth, CO2 emissions, and technological advancement (Manigandan et al., 2023) [45]. The role of scientific research in advancing technology while addressing these challenges is critical, as emphasized by Imaz and Sheinbaum (2017) [46].

In this context, studying countries’ scientific productivity and priorities regarding technologies that contribute to Sustainable Development Goals (SDGs) is crucial for several reasons. First, scientific productivity directly influences a country’s ability to develop and deploy technologies that can address SDG-related challenges. Countries with high scientific output in relevant fields are better positioned to innovate and implement solutions that enhance sustainability, improve health outcomes, and mitigate environmental impacts. Understanding the differentiated productivity and priorities among countries allows us to identify which nations are leading in specific technological advancements and which may be lagging behind, providing a basis for targeted international collaboration and support (Adel and Alani, 2024) [6]. Additionally, the profiles of scientific priorities reflect the unique needs and strategic goals of different countries. For instance, a nation heavily reliant on agriculture might prioritize research in AI and IoT for smart farming, while another with a focus on urbanization might invest more in technologies for smart cities. By studying these differentiated priorities, we can better understand how countries are aligning their scientific efforts with their most pressing developmental needs, which is essential for ensuring that the right technologies are developed and deployed where they are needed most (Palomares et al., 2021) [21].

Moreover, examining the factors that shape scientific productivity—such as funding availability, educational infrastructure, and government policies—can reveal systemic strengths and weaknesses in how countries support technological innovation (Rughiniș et al., 2024) [47]. For example, disparities in funding and infrastructure often lead to significant gaps in research output between developed and developing countries (Bran et al., 2024) [48] which can exacerbate the digital divide and hinder global progress toward achieving the SDGs (Managi et al., 2021; Tjoa and Tjoa, 2016; Varriale et al., 2024) [49,50,51]. Identifying these gaps allows for more effective interventions, such as capacity-building initiatives and policy reforms, to boost scientific productivity where it is most needed (Imaz & Sheinbaum, 2017) [46]. Finally, understanding the global landscape of scientific productivity and priorities helps to anticipate and mitigate the risks associated with technological deployment. For instance, countries with low scientific output in cybersecurity may be more vulnerable to cyber attacks as they adopt new digital technologies, posing risks to their infrastructure and citizens. By recognizing these vulnerabilities, international efforts can be better coordinated to provide the necessary support and resources to countries that might otherwise face significant challenges in safely and effectively implementing technologies that contribute to the SDGs (Wu et al., 2018) [20].

Scientific output and productivity on a global scale are influenced by economic, institutional, and societal factors. Economic wealth, particularly as measured by GDP or GNI, is consistently identified as a significant predictor of research output across various studies (Rodríguez-Navarro and Brito, 2023; Onyancha, 2022; Lancho-Barrantes et al., 2023; Zhang et al., 2022; Jamjoom and Jamjoom, 2022; Rahman and Fukui, 2022) [52,53,54,55,56,57]. Research and development (R&D) expenditure, both as a percentage of GDP and in absolute terms, also plays a great role in driving scientific productivity (Lancho-Barrantes et al., 2023; Rahman and Fukui, 2022) [54,57]. However, the efficiency of a country’s research system is important, especially for producing high-impact research (Rodríguez-Navarro and Brito, 2023) [52]. Institutional factors, such as the presence and quality of academic and research institutions, also contribute to scientific output (Lancho-Barrantes et al., 2023; Jamjoom and Jamjoom, 2022) [54,56]. The number of universities ranked in the world’s top 500 and the number of journals in specific fields correlate strongly with productivity (Jamjoom and Jamjoom, 2022) [56]. Additionally, factors like international collaboration, access to resources, and library support have been identified as important drivers of scientific productivity (Szuflita-Żurawska and Basińska, 2022; Wahid et al., 2023) [58,59].

Societal and policy-related factors also play a significant role in shaping scientific output. Good governance, absence of a communist past, and smaller population size have been associated with higher-quality scientific publications (Allik et al., 2023) [60]. Language factors, particularly the prevalence of English, can influence both productivity and citation rates (Dragos and Dragos, 2022; Tasli et al., 2022; Gantman, 2022) [61,62,63]. Environmental policies, as measured by the Environmental Performance Index, have been shown to predict research output in environmental sciences (Dragos and Dragos, 2022) [61]. Interestingly, when controlling for societal factors, some studies have found that economic indicators become less significant in predicting scientific excellence (Allik et al., 2023) [60]. At the individual level, factors such as time management, academic rank, and qualifications significantly affect productivity (Wahid et al., 2023) [59]. It is worth noting that while overall scientific output is important, the quality and impact of research, as measured by citations and representation in top-cited papers, are crucial indicators of a country’s scientific contribution (Rodríguez-Navarro and Brito, 2023; Tasli et al., 2022; Doi et al., 2022) [52,62,64]. These findings collectively suggest that while economic resources provide a foundation for scientific productivity, the efficiency of research systems, institutional quality, and societal factors also play important roles in shaping a country’s scientific output and impact.

The purpose of our research is to provide an analysis of the global landscape of technological innovation in support of Sustainable Development Goals (SDGs). By examining the distribution of research productivity across 18 key technologies relevant to SDGs, identifying country-level factors that predict this productivity, and exploring differences in national technological focus and priorities, this study aims to offer relevant information for policymakers, researchers, and stakeholders involved in sustainable development efforts. The findings will contribute to a better understanding of how technological innovation is being leveraged worldwide to address pressing global challenges. They will also highlight potential gaps or disparities in research efforts, while informing strategic decision-making related to resource allocation and international collaboration for advancing SDGs through technological means. Additionally, this research seeks to establish a foundation for future studies by mapping the current state of SDG-related technological innovation and identifying areas that require further investigation or support.

Thus, to understand the global landscape of technological innovation in support of Sustainable Development Goals (SDGs), we explore three key research questions. First, what is the global distribution of research productivity concerning the selected 18 technologies that are highly relevant to SDGs? These technologies include Artificial Intelligence, Robotics Automation, renewable energy, Internet of Things, Geographic Information Systems, Big Data Analytics, Cybersecurity Protection, Biotechnology Genetics, Augmented and Virtual Reality, 3D printing, next-generation networks, biometric identity and cybersecurity, cloud computing, drones and UAVs, Natural Language Processing, Telemedicine and Healthcare, blockchain technology, and digital twin simulation. Secondly, what factors at the country level predict this productivity? Finally, how do countries differ in terms of their focus and priorities among these selected technologies?

2. Materials and Methods

This study employs a quantitative approach to analyze the relationship between countries’ research priorities in Sustainable Development Goal (SDG)-relevant technologies and various socio-economic indicators. The research process involved data collection, preprocessing, and multiple analytical techniques.

We began by conducting a comprehensive review of reports from organizations such as the United Nations Development Programme (UNDP) and the World Economic Forum (WEF) to identify digital technologies that contribute to the SDGs. This review, discussed in the Introduction section, resulted in the selection of 18 technological domains, including those indirectly connected to the digital sphere. This approach ensures that our analysis is grounded in the current global discourse on technology’s role in achieving the SDGs. The selection process prioritized technologies that are frequently cited in these reports as having significant potential to contribute to multiple SDGs. For instance, Artificial Intelligence and Big Data Analytics were included due to their cross-cutting applications in areas such as healthcare, education, and environmental management. Renewable energy technologies were selected for their critical role in addressing climate change (SDG 13) and ensuring access to clean energy (SDG 7). We also aimed to include a mix of established and emerging technologies. While some domains like Robotics Automation and Geographic Information Systems represent more mature fields with proven SDG applications, others like blockchain technology and digital twins represent emerging areas with promising potential for sustainable development. The selection also considered the breadth of SDG coverage. For example, Telemedicine and Healthcare technologies directly relate to SDG 3 (Good Health and Well-being), while Internet of Things and smart city technologies can contribute to multiple goals including SDG 11 (Sustainable Cities and Communities) and SDG 9 (Industry, Innovation, and Infrastructure). We acknowledge that this selection is not exhaustive and may not capture every niche or emerging technology relevant to the SDGs. The rapidly evolving nature of technological innovation means that new relevant technologies may emerge over time. However, we believe that our selection provides a robust representation of the key technological domains currently recognized as crucial for SDG achievement. The 18 selected domains offer a balance between breadth of coverage and analytical feasibility. This number allows for a broad analysis while still maintaining a manageable scope for cross-country comparisons and in-depth examination of research priorities. Future studies might benefit from expanding this list or focusing on more specialized technological niches. However, for the purposes of this global comparative analysis, we believe these 18 domains provide a good foundation for understanding the landscape of SDG-relevant technological research across countries.

For each identified technological domain, we constructed specific queries to extract total publication data from the Web of Science (WoS) database (Clarivate, 2024) [65], as shown in

Table 1. Queries used to extract publication data from Web of Science. Source: Authors’ analysis.

Variable	WoS Query	Total No. of Publications
Artificial Intelligence	“Artificial Intelligence” OR “Machine Learning” OR “Deep Learning”	754,327
Robotics Automation	“Robotics” OR “Automation” OR “Robotic Systems”	218,274
Renewable Energy	“Renewable Energy” OR “Green Energy” OR “Clean Energy”	184,682
Internet Of Things	“Internet of Things” OR “IoT” OR “Connected Devices”	140,073
Geographic Information Systems	“Geographic Information Systems” OR “GIS” OR “Geospatial Analysis” OR “Spatial Data”	133,690
Big Data Analytics	“Big Data” OR “Data Analytics” OR “Data Science”	124,311
Cybersecurity Protection	“Cybersecurity” OR “Cyber-security” OR “Cyber security” OR “Intrusion detection” OR “Phishing” OR “Malware” OR “Spyware” OR “Ransomware” OR “Cyber attack” OR “Adware” OR “Hacking” OR “Network Security” OR “Cryptography”	121,864
Biotechnology Genetics	“Biotechnology” OR “Genetic Engineering” OR “Life Sciences” OR “Synthetic Biology”	117,407
Augmented and Virtual Reality	“Augmented Reality” OR “Virtual Reality” OR “Mixed Reality”	104,317
Three D Printing	“3D Printing” OR “Additive Manufacturing”	94,484
Next-Gen Networks	“5G” OR “6G”	84,476
Biometric Identity and Cybersecurity	“Biometric Identification” OR “Biometrics” OR “Digital ID” OR “Biometric Authentication” OR “Biometric Verification” OR “Facial Recognition” OR “Fingerprint Scanning” OR “Iris Recognition” OR “Voice Recognition” OR “DNA Fingerprinting” OR “Retina Scanning” OR “Hand Geometry” OR “Behavioral Biometrics” OR “Multimodal Biometrics” OR “eID” OR “Electronic Identity” OR “Identity Management Systems” OR “Identity Verification” OR “Secure Identification” OR “Digital Identity Platforms” OR “Identity Access Management” OR “Identity Solutions” OR “Biometric Security” OR “Mobile Identity Verification” OR “Blockchain Identity Management” OR “Smart Identity Cards” OR “Remote Biometric Identification” OR “Biometric Data Protection” OR “Digital Certificates” OR “Federated Identity Management” OR “Passwordless Authentication” OR “RFID” OR “Radio-Frequency Identification”	71,984
Cloud Computing	“Cloud Computing” OR “Cloud Services”	71,451
Drones and UAVs	“Drones” OR “Unmanned Aerial Vehicles” OR “Autonomous Vehicles”	60,935
Natural Language Processing	“Natural Language Processing” OR “NLP” OR “Language Technology”	56,021
Telemedicine and Healthcare	“Telemedicine” OR “Telehealth” OR “Remote Healthcare”	55,399
Blockchain Technology	“Blockchain” OR “Distributed Ledger”	37,462
Digital Twin Simulation	“Digital Twins” OR “Virtual Twins” OR “Simulation Models”	33,563

below. These queries were designed to capture the number of publications for each country on a given SDG-relevant technological topic. The data were downloaded from Web of Science between 6 and 9 August 2024.

We thus rely on publication data from the Web of Science (WoS), a leading database for scientific research. While this choice provides access to a vast collection of high-quality, peer-reviewed publications, it is important to acknowledge the limitations this imposes on our analysis, particularly regarding the representation of non-English research. WoS primarily indexes English-language journals, which leads to an underrepresentation of research published in other languages. This bias could particularly affect our analysis of SDG-relevant technological research from non-English-speaking countries, especially those in the Global South where English may not be the primary language of scientific communication. The decision to use WoS was based on several factors. First, it offers comprehensive coverage of international, high-impact journals across various scientific disciplines. Second, it provides standardized metadata that facilitate large-scale bibliometric analysis. Third, its focus on peer-reviewed publications ensures a baseline of scientific quality. However, this approach does not capture the full spectrum of global research efforts related to SDGs. Research published in local languages, in regional journals not indexed by WoS, or disseminated through non-traditional channels (such as policy papers or technical reports) is underrepresented in our dataset. To mitigate this limitation, future studies could consider incorporating data from additional sources. These might include regional databases such as SciELO for Latin American research, or databases with broader language coverage like Google Scholar. Additionally, manual searches for non-English publications in specific countries or regions could provide a broader view of the research landscape. It is also worth noting that the bias towards English-language publications in WoS may reflect broader patterns in global scientific communication, where English often serves as a lingua franca. While this facilitates international knowledge exchange, it also presents challenges for non-English-speaking researchers and potentially skews our understanding of global research efforts. Thus, the patterns and trends we identify are most representative of research that has achieved international visibility in English-language, WoS-indexed journals.

We use publication counts from the Web of Science as a proxy measure of research productivity in SDG-relevant technologies. While this approach provides relevant information about the volume and distribution of research efforts across countries and technological domains, it has both advantages and limitations. Publication counts are quantifiable and widely available and allow for comparisons across different countries and fields of study. They also reflect the academic community’s engagement with specific research areas, which can indicate trends or priorities in technological development. However, publication counts alone do not capture the full spectrum of a country’s contribution to SDG-relevant technological innovation. They do not directly measure the quality, impact, or practical applications of the research, which are important factors in assessing how effectively scientific output translates into sustainable development outcomes. To provide a broader picture, future studies could incorporate additional metrics. Citation counts could serve as a proxy for the impact and influence of published research within the scientific community. Patent data could offer insights into the commercialization potential and practical applications of technological innovations. Altmetrics, which measure the attention research receives in non-traditional channels such as social media and policy documents, could indicate broader societal impact and relevance to SDG implementation. Furthermore, case studies of successful technology deployments, analysis of research collaboration networks, and assessment of technology transfer activities could enrich our understanding of how research output contributes to real-world sustainable development challenges. Incorporating data on research funding allocation and national science policies could also provide context for interpreting publication patterns.

It is important to note that while these additional metrics would offer a clearer view of research activity, they also present challenges in terms of data availability, consistency across countries, and methodological complexity. Our current approach using publication counts should be viewed as a starting point for understanding global patterns in SDG-relevant technological research, laying the groundwork for more detailed analyses incorporating multiple dimensions of research impact and application.

The collected data were organized in IBM SPSS software version 29.0.2.0, with countries treated as cases. The database includes the number of publications for each of the 18 technological domains, the scientific productivity measured as the number of publications in each domain normalized by the country’s population (per 1000 inhabitants), and the decreasing rank order of the scientific productivity for each domain.

In addition to the research productivity data, our dataset incorporates the Augmented Human Development Index (AHDI) values and its component indicators (de la Escosura, 2024) [66]. The AHDI is an extension of the United Nations Development Programme’s (UNDP) Human Development Index (HDI), providing a more comprehensive measure of human development and well-being across countries. Our dataset includes key AHDI indicators: life expectancy at birth, representing the public health dimension; mean years of schooling, capturing the education dimension; Gross National Income (GNI) per capita, measuring the standard of living; and the liberal democracy index (Herre, 2024) [67], which reflects the level of democratic governance and civil liberties in a country. We also include in our dataset information on research funding as part of GDP, at country level, as well as countries’ SDG score. Descriptive information is available in

Table 2. Descriptive values for variables included in the study. Source: Authors’ analysis.

Type	Indicator	N	Min.	Max.	Mean	Std. Deviation
Socio-demographic indicators	AHDI	162	0.12	0.89	0.45	0.20
	Life expectancy	195	52.53	85.95	71.28	7.75
	Average years of schooling	191	2.11	14.09	8.99	3.17
	GNI per capita	193	731.79	146,829.70	20,136.39	21,756.09
	Liberal democracy index	176	0.01	0.89	0.39	0.26
	Research spending per GDP	150	0.01	5.56	0.83	1.02
	SDG index	167	40.14	86.35	67.63	10.08
Research productivity (country publications per population)	Artificial Intelligence	190	0.00	8.99	0.26	0.73
	Robotics Automation	174	0.00	1.28	0.07	0.14
	Renewable Energy	188	0.00	0.62	0.07	0.11
	Internet of Things	169	0.00	0.49	0.05	0.08
	Geographic Information Systems	191	0.00	0.53	0.04	0.06
	Big Data Analytics	183	0.00	1.28	0.05	0.12
	Cybersecurity	164	0.00	0.54	0.04	0.07
	Biotechnology Genetics	174	0.00	0.31	0.03	0.05
	Augmented and Virtual Reality	159	0.00	0.46	0.04	0.07
	3D Printing	148	0.00	0.40	0.03	0.05
	Next-Gen Networks	169	0.00	0.49	0.03	0.06
	Biometric Identity and Cybersecurity	178	0.00	0.26	0.02	0.03
	Cloud Computing	156	0.00	0.22	0.02	0.03
	Drones and UAVs	160	0.00	0.79	0.03	0.07
	Natural Language Processing	151	0.00	0.20	0.02	0.03
	Telemedicine and Healthcare	190	0.00	0.15	0.02	0.03
	Blockchain Technology	155	0.00	0.41	0.02	0.04
	Digital Twin Simulation	157	0.00	0.22	0.01	0.03
Within-country ranking (1 = first, 18 = last) of research productivity	Artificial Intelligence	120	1.00	2.00	1.06	0.24
	Robotics Automation	120	2.00	13.00	4.87	2.68
	Renewable Energy	120	1.00	10.00	3.13	1.65
	Internet of Things	120	2.00	17.00	5.38	2.75
	Geographic Information Systems	120	2.00	16.00	5.23	3.04
	Big Data Analytics	120	3.00	16.00	6.70	1.93
	Cybersecurity	120	2.00	17.00	7.48	3.18
	Biotechnology Genetics	120	2.00	18.00	8.91	4.29
	Augmented and Virtual Reality	120	3.00	18.00	11.04	4.08
	3D Printing	120	3.00	18.00	11.99	3.52
	Next-Gen Networks	120	2.00	18.00	10.37	3.55
	Biometric Identity and Cybersecurity	120	4.00	18.00	11.86	2.94
	Cloud Computing	120	2.00	18.00	11.50	3.67
	Drones and UAVs	120	8.00	18.00	13.29	2.28
	Natural Language Processing	120	2.00	18.00	13.17	2.96
	Telemedicine and Healthcare	120	4.00	18.00	13.18	3.85
	Blockchain Technology	120	5.00	18.00	14.34	3.44
	Digital Twin Simulation	120	3.00	18.00	16.02	2.67

below. The total number of countries included in the regression analysis is 135, representing those nations for which both total research productivity data and socio-demographic indicators are available. The cluster analysis encompasses 120 countries, specifically those that have published research across all 18 technological fields under consideration.

Our analytical methods include several approaches. We conducted initial exploratory data analysis to understand the distribution of research productivity across countries and domains, calculating Pearson correlation coefficients to examine relationships between variables (see the Supplementary Materials). A multiple linear regression model was estimated to predict total SDG-related research productivity based on socio-economic indicators discussed above. To identify patterns in research priorities across countries, we performed a K-means cluster analysis based on the scientific productivity data for the 18 technological domains. This method grouped countries with similar research focus patterns, allowing for the identification of distinct approaches to SDG-related technological research. We also created geographical visualizations with the help of Office 365 Excel to illustrate the spatial distribution of research productivity and cluster memberships, aiding in the identification of regional patterns and disparities.

3. Results

Scientific productivity levels across countries for the selected SDG-relevant technologies exhibit strong intercorrelations, with most Bravais–Pearson correlation coefficients ranging between 0.8 and 0.9 (see Section S2 of the Supplementary Materials for detailed values). A principal component analysis of the 18 productivity variables reveals that the first component accounts for about 79% of the total variance, while the second component explains only 7.3% (full results in Section S3 of the Supplementary Materials). This high degree of commonality among the variables suggests that examining the variability in total SDG-related productivity, calculated as the sum of all individual productivities, is a valid and informative first step in our analysis. This approach provides an initial overview of overall research output in SDG-relevant fields across countries, setting the foundation for more detailed exploration of specific patterns and variations.

The global distribution of total scientific productivity on SDG-relevant technologies, as illustrated by the map in Figure 1 below and accompanying data in Supplementary Materials, reveals a landscape characterized by significant disparities across countries and regions. This distribution is highly skewed, with a small number of nations, predominantly in Europe and North America, producing a disproportionately large amount of research relative to their population. Switzerland, Denmark, Norway, and Sweden emerge as leaders in Europe, showcasing a very high research output per capita. Australia stands out as a major outlier in the Asia–Pacific region, potentially due to its well-funded research institutions and focus on sustainability issues. Among larger countries, the United Kingdom, Canada, and Germany demonstrate high productivity, reflecting substantial investments in research and commitment to Sustainable Development Goals.

The United States, while undoubtedly a global research powerhouse, ranks lower in per capita terms for SDG-relevant technologies compared to many European countries. This positioning might be attributed to its large population and a broader distribution of research interests across various fields, including defense, space exploration, and biomedical research, which may not directly fall under SDG categories.

In Asia, Singapore, South Korea, and Japan exhibit relatively high SDG-relevant productivity, highlighting their strong research capabilities in technological solutions to sustainability challenges. China, despite its growing global influence in research, shows a lower per capita output, likely due to its massive population and possibly a research focus that extends beyond SDG-specific technologies.

Middle Eastern countries like Qatar, Saudi Arabia, and the United Arab Emirates demonstrate moderate levels of productivity, potentially reflecting recent investments in research and development, although their focus may include energy technologies not all classified under SDG frameworks. Latin America presents a varied picture, with Chile standing out with higher SDG productivity compared to its neighbors. Brazil and Argentina, despite being regional powers, show lower per capita output in SDG-relevant technologies, which may indicate different research priorities or challenges in research infrastructure.

Russia’s position is notably lower than might be expected given its scientific legacy, possibly reflecting shifts in research priorities, funding allocations, or a focus on areas like space technology and nuclear physics that may not directly align with SDG classifications. India, a major emerging economy with a growing research sector, shows relatively low per capita output in SDG technologies. This could be attributed to its vast population, uneven distribution of research capabilities, and potentially a prioritization of research in areas deemed more immediately crucial for its development.

African countries generally exhibit lower productivity rates in SDG-relevant technologies, with South Africa being a notable exception. This trend reflects the continent’s ongoing challenges in research infrastructure and funding, as well as possible divergent research priorities.

3.1. Linear Regression Modeling

Figure 2 below illustrates the relationship between a country’s research productivity on Sustainable Development Goal (SDG) technologies and its Augmented Human Development Index (AHDI). It presents two scatterplots, the left showing raw data and the right displaying log-transformed data for both variables. The raw data plot reveals a non-linear relationship between AHDI and SDG productivity. In contrast, the log-transformed plot shows a relationship that more closely approximates linearity. This transformation results in a more even distribution of data points along a diagonal line, indicating a stronger linear correlation between log(AHDI) and log(SDG productivity). The log transformation proves particularly useful for linear modeling, especially when using a linear regression model to analyze SDG technologies as a function of AHDI indicators. This approach addresses several key aspects of data analysis. It enhances linearity, which is crucial for linear regression models. The transformation also improves homoscedasticity by creating a more consistent spread of data points across the range of log(AHDI) values. Additionally, it can help normalize skewed data. From an interpretative standpoint, the log–log model allows coefficients to be understood as elasticities, representing percentage changes rather than absolute changes. This transformation also effectively manages scale issues by bringing large values closer together and spreading out smaller values, facilitating analysis across a wide range of values.

This linear regression model in

Table 3. Linear regression model of total country research productivity (publications/population) in SDG technologies (log transformation) as a function of AHDI indicators, research spending as proportion of GDP, and the SDG score (all with log transformation). N = 135 countries. Source: Authors’ analysis.

	Unstandardized Coefficients		Standardized Coefficients	t	Sig.	Collinearity Statistics
	B	Std. Error	Beta			Tolerance	VIF
(Constant)	−9.127	1.744		−5.234	0.000
log life expectancy UNDP	2.764	1.105	0.146	2.502	0.014	0.231	4.338
log average years of schooling	−0.062	0.276	−0.013	−0.224	0.823	0.246	4.058
log GNI per capita UNDP	1.149	0.126	0.615	9.110	0.000	0.171	5.839
log liberal democracy index	0.251	0.074	0.112	3.384	0.001	0.717	1.394
log research spending/GDP	0.408	0.062	0.262	6.627	0.000	0.499	2.003
log SDG score	−0.627	0.983	−0.042	−0.637	0.525	0.184	5.433

Dependent variable: Log of total country research productivity (publications/population) in SDG technologies.

below examines the relationship between a country’s research productivity in SDG-related technologies and several predictors derived from AHDI indicators, along with research funding as a proportion of GDP and the country-specific SDG score. The model uses log-transformed variables, allowing for a more linear relationship between the predictors and the outcome. The standardized beta coefficients provide information about the relative importance of each predictor in explaining variations in research productivity. These coefficients are directly comparable as they are measured on the same scale.

The most influential predictor in this model is log GNI per capita (UNDP), with a beta coefficient of 0.615 and a highly significant p-value (p < 0.001). This suggests that a country’s economic output has the strongest positive relationship with its research productivity in SDG technologies. Research spending as a proportion of GDP is the second most important predictor, with a beta coefficient of 0.262 (p < 0.001). This indicates that countries allocating a higher percentage of their GDP to research tend to have higher research productivity in SDG-related technologies, which is a logical and expected relationship.

Life expectancy shows a positive relationship (beta = 0.146, p = 0.014) with research productivity, suggesting that countries with better health outcomes also tend to produce more research on SDG technologies. This could reflect overall development levels or potentially a greater focus on sustainability in countries with higher life expectancy, as life expectancy is a good proxy for broader levels of societal development (Rughiniș et al., 2022) [68]. The liberal democracy index also shows a positive relationship (beta = 0.112, p = 0.001), indicating that more democratic countries tend to have slightly higher research productivity in SDG technologies. This could be due to factors such as academic freedom or societal priorities in democratic nations.

Interestingly, average years of schooling and SDG score coefficients are very small and not statistically significant (p > 0.05). This suggests that, after accounting for the other factors in the model, these variables do not have a reliable relationship with research productivity in SDG technologies.

The model’s very strong predictive power, as indicated by an R Square of 0.900 and an Adjusted R Square of 0.895, demonstrates its ability to explain the variance in research productivity in SDG-related technologies across countries. Such a high level of predictive power suggests that the model has captured key factors influencing a country’s research output in SDG technologies. It implies that the combination of economic indicators (GNI per capita and research spending), health outcomes as possible proxies of development (life expectancy), and governance factors (democracy index) provides a good framework for understanding variations in research productivity across nations. However, it is important to note that while this model shows strong predictive capabilities, it does not necessarily imply causation, and there may be reciprocal relationships and other unmeasured factors or interactions at play.

3.2. Classification Analysis

While examining overall scientific productivity in SDG-relevant technologies provides valuable information, it does not fully capture the landscape of research focus across different countries. Countries may have similar levels of total productivity but different priorities in terms of which technologies they emphasize in their research efforts. Understanding these priorities is important as it reveals strategic national focuses in technological development, helps identify potential areas of expertise or specialization for each country, can highlight gaps or underexplored areas in a country’s research landscape, and facilitates more targeted international collaboration based on complementary strengths.

To capture these research priorities, we propose using a within-country rank metric for each of the 18 SDG-relevant technologies. This metric assigns a rank from 1 to 18 for each technology within each country, where 1 represents the most intensely studied technology and 18 the least. The ranking is based on the relative productivity of each technology within the country, regardless of its absolute productivity level compared to other countries.

The use of ranks in our classification analysis offers several methodological advantages. Primarily, it normalizes data across countries with varying research outputs, enabling meaningful comparisons of technological priorities irrespective of absolute productivity levels. This approach captures distinct information about research focus, as evidenced by the low bivariate correlations among ranking indicators (with most absolute values around 0.2–0.3; see details in Section S4 of the Supplementary Materials), in contrast to the high intercorrelations observed in raw productivity measures. By shifting focus to relative priorities within each country, we can discern strategic national emphases in technological development that might otherwise be obscured by overall output differences. The simplicity of ranks enhances interpretability, providing a clear metric for assessing the relative importance of technologies within a country’s research landscape. Furthermore, this method effectively addresses the issue of multicollinearity present in relative productivity variables, reducing redundancy in the classification process. While rank differences are uniform, this approach facilitates the identification of broad patterns in technological priorities across countries and regions, which aligns with our primary analytical objective. It is important to acknowledge that underlying productivity differences may vary more than ranks suggest, and future research could explore methods that retain more granular information while still enabling cross-country comparisons.

Using this metric, we performed a K-means cluster analysis to classify countries based on their shared research priorities. This approach groups countries with similar patterns of technological focus, regardless of their absolute levels of productivity. Such an analysis reveals common patterns of research prioritization among countries with similar economic structures, geographic features, or development levels. It uncovers unexpected similarities in research focus between seemingly disparate countries, potentially indicating shared challenges or strategic alignments. The analysis also highlights distinct approaches to addressing SDGs through technological research, which could inform policy decisions and international cooperation strategies. Additionally, it reveals potential gaps in global research efforts, where certain technologies are consistently deprioritized across multiple country clusters.

We opted for a five-cluster solution for the K-means analysis of research priorities among 18 SDG-relevant technologies, which offers a good balance between detail and interpretability. The choice of a five-cluster solution for our analysis was driven by the goal of maximizing analytical nuance while maintaining a robust typological approach. We aimed to identify the highest number of clusters that would not result in groups that were too small for meaningful interpretation. Comparing the five-cluster and six-cluster solutions revealed that the latter produced two problematic groupings: one with only three cases and another with a single case. Such small clusters are typically difficult to interpret as distinct types and may represent outliers rather than meaningful groupings. In contrast, the five-cluster solution provided a more balanced distribution, with each cluster containing at least 11 cases. This approach provides an in-depth view of the global research landscape, revealing distinct groups of countries with similar priorities while avoiding over-segmentation. The balanced distribution of countries across the five clusters indicates that the solution has captured meaningful differences in research focus, enhancing the robustness of the analysis.

3.2.1. Cluster Profiles

Across all clusters, Artificial Intelligence consistently ranks as the top research priority, with an average rank of 1.1, indicating its pervasive importance in SDG-related research globally. Other two technologies are also relatively similarly ranked in all clusters. Big Data Analytics and drones/UAVs show less prioritization, with average ranks of 6.7 and 13.3, respectively (see

Table 4. Final cluster centers for K-means analysis of countries’ research focus on SDG technologies. Smaller values indicate higher priorities. Source: Authors’ analysis.

	1	2	3	4	5
	Eco-Tech Innovators	Cyber- Digital Architects	Bio- Industrial Pioneers	Geo-Data Security Analysts	Cyber- Sustainable Integrators
Artificial Intelligence	1.00	1.00	1.03	1.14	1.08
Robotics Automation	3.20	4.82	2.42	6.00	7.22
Renewable Energy	2.40	4.27	3.84	3.18	2.53
Internet Of Things	5.10	3.45	6.94	7.55	3.44
Geographic Information Systems	4.55	11.64	5.52	2.77	4.89
Big Data Analytics	6.45	5.64	6.77	6.86	7.00
Cybersecurity Protection	7.05	5.55	9.39	9.64	5.36
Biotechnology Genetics	9.90	13.09	5.03	5.64	12.42
Augmented and Virtual Reality	8.40	8.55	7.68	13.41	14.72
Three D Printing	11.00	10.45	8.77	14.27	14.39
Next-Gen Networks	12.90	6.45	11.42	11.05	8.83
Biometric Identity and Cybersecurity	13.25	13.82	12.58	9.27	11.44
Cloud Computing	12.15	11.18	14.77	13.05	7.47
Drones and UAVs	12.75	11.82	13.26	13.82	13.75
Natural Language Processing	13.70	14.09	14.10	12.55	12.17
Telemedicine and Healthcare	14.35	16.27	14.48	7.55	13.89
Blockchain Technology	14.55	11.55	17.10	15.95	11.72
Digital Twin Simulation	17.40	16.82	15.58	13.00	17.22
Total number of countries a N = 120	20	11	31	22	36

^a. The numbers indicate the total number of countries found in each cluster.

and

Table 5. Highest research priorities for the five clusters. Smaller values indicate higher priorities. Source: Authors’ analysis.

	1	2	3	4	5
	Eco-Tech Innovators	Cyber- Digital Architects	Bio- Industrial Pioneers	Geo-Data Security Analysts	Cyber- Sustainable Integrators
Artificial Intelligence
Robotics Automation	3.20		2.42
Renewable Energy	2.40				2.53
Internet Of Things		3.45			3.44
Geographic Information Systems	4.55			2.77
Big Data Analytics
Cybersecurity Protection		5.55			5.36
Biotechnology Genetics			5.03	5.64
Augmented and Virtual Reality	8.40	8.55	7.68
Three D Printing			8.77
Next-Gen Networks		6.45			8.83
Biometric Identity and Cybersecurity				9.27
Cloud Computing					7.47
Drones and UAVs
Natural Language Processing				12.55	12.17
Telemedicine and Healthcare				7.55
Blockchain Technology		11.55			11.72
Digital Twin Simulation				13.00

below), suggesting a more specialized or emerging focus in these areas.

In Table 5, we have selected and highlighted the clusters with the highest priority for all SDG technologies included in analysis. For Artificial Intelligence, Big Data Analytics, and drones and UAVs, cluster centers were very similar and did not indicate selective priorities.

Cluster 1, the “Eco-Tech Innovators”, prioritizes renewable energy research, followed closely by robotics and automation. These countries also demonstrate a stronger focus on Geographic Information Systems and Augmented/Virtual Reality compared to other clusters. This suggests a research landscape oriented towards sustainable energy solutions and advanced technological applications for environmental management.

Cluster 2, the “Cyber-Digital Architects”, emphasizes Internet of Things research, with secondary focuses on cybersecurity and next-generation networks. These countries also show relatively higher interest in blockchain technology, indicating a research agenda centered on building and securing comprehensive digital infrastructures.

Cluster 3, the “Bio-Industrial Pioneers”, prioritizes robotics and automation research, followed by biotechnology and genetics. This cluster also shows heightened interest in Augmented/Virtual Reality and 3D printing, displaying a research focus on the integration of advanced manufacturing techniques with biotechnology applications.

Cluster 4, the “Geo-Data Security Analysts”, demonstrates a strong emphasis on Geographic Information Systems. Notably, this group shows more interest in biometric identity and cybersecurity, Natural Language Processing, and digital twin simulation compared to other clusters, although these remain lower priorities overall. This pattern indicates a research landscape focused on geospatial technologies with emerging interests in advanced computing and cognitive systems.

Cluster 5, the “Cyber-Sustainable Integrators”, prioritizes renewable energy research, closely followed by the Internet of Things. These countries also have a relatively strong focus on Cybersecurity Protection and cloud computing. This shows a research agenda balancing sustainable energy development with digital infrastructure and security, aiming to integrate sustainable physical systems with advanced cyber technologies.

Across all clusters, the consistent high ranking of Artificial Intelligence underscores its central role in SDG-related research globally. The varying priorities in other technologies reflect diverse approaches to SDG challenges among different groups of countries. These distinct research foci could be influenced by factors such as economic structures, technological capabilities, or specific SDG priorities. Understanding these patterns can inform international collaboration strategies, guide policy decisions, and highlight potential areas for knowledge transfer between countries with different research strengths. Moreover, the generally lower rankings of technologies like Natural Language Processing and digital twin simulation across clusters may indicate emerging research areas with potential for growth and increased relevance to SDG achievement in the future.

Table 6. Socio-demographic profile of clusters regarding the main predictors of SDG research productivity. Source: Authors’ analysis.

	1 Eco-Tech Innovators	2 Cyber- Digital Architects	3 Bio- Industrial Pioneers	4 Geo- Data Security Analysts	5 Cyber- Sustainable Integrators
	Mean	Mean	Mean	Mean	Mean
Life expectancy	74.96	81.38	78.15	69.22	71.04
GNI per capita (thou. USD)	26.45	69.72	38.07	12.40	17.71
Liberal democracy index	0.52	0.51	0.65	0.39	0.22
Research spending/GDP	0.84	1.97	1.95	0.46	0.39
SDG score	75.05	74.43	78.40	66.63	65.30

shows clusters’ average profile concerning the main predictors of SDG research productivity, according to the regression model discussed in the section above. The variations in indicator values between clusters highlight global disparities in economic development, democratic freedoms, and research capacity. However, these differences do not necessarily reflect variations in technological sophistication or innovation potential. Each cluster, regardless of its socio-economic profile, is pursuing advanced technologies aligned with its specific developmental needs and constraints.

Cluster 1, the Eco-Tech Innovators, shows moderate values across all indicators. The life expectancy and GNI per capita are slightly above the global average, reflecting a mix of developed and developing economies. Countries like Australia and Canada represent the higher end of these indicators, while nations such as Indonesia and Turkey represent the lower end. The moderate liberal democracy index suggests reasonably open societies, with Estonia exemplifying a highly democratic system within this cluster. The relatively low research spending as a percentage of GDP might be illustrated by countries like Greece or Colombia, which have more limited research budgets compared to top global spenders.

Cluster 2, the Cyber-Digital Architects, stands out with the highest life expectancy and GNI per capita among all clusters. Singapore and South Korea are prime examples of countries in this cluster with high life expectancy and strong economies. The cluster’s high research spending is reflected in the robust R&D investments of countries like Finland and China. Interestingly, the liberal democracy index is lower than might be expected given the economic indicators, with Singapore and China representing more controlled political systems, while Ireland exemplifies a more open democracy within this group.

Cluster 3, the Bio-Industrial Pioneers, shows a balanced profile of high development. Its liberal democracy index is the highest among all clusters, with countries like Sweden and the Netherlands representing strong democratic traditions. The high SDG score is reflected in the sustainable development efforts of nations like Denmark and Germany. Research spending is also high, with Japan and the United States illustrating significant investments in R&D.

Cluster 4, the Geo-Data Security Analysts, shows lower values across all indicators compared to the global average. The life expectancy and GNI per capita reflect the cluster’s composition of primarily middle-income countries. Argentina and the Philippines represent the range of economic development within this cluster. The lower research spending could be exemplified by countries like Kenya or Cuba, which have more limited resources for R&D but are strategically focusing on specific technological areas.

Cluster 5, the Cyber-Sustainable Integrators, shows the lowest values for the liberal democracy index and research spending. Countries like Saudi Arabia and Egypt illustrate the more controlled political systems prevalent in this cluster. However, the GNI per capita is higher than Cluster 4, with the United Arab Emirates representing the higher end of the economic spectrum and India the lower end. Despite lower research spending, countries in this cluster like Malaysia are making strategic investments in sustainable and digital technologies.

3.2.2. Regional Specificities

The distribution of countries across the five clusters reveals interesting patterns in research priorities for SDG-relevant technologies, according to Figure 3 below. These commonalities and differences suggest that while all clusters prioritize technology, the specific focus areas are influenced by each country’s economic structure, natural resources, developmental challenges, and long-term strategic goals. The clustering also indicates that countries with historically different economic models or even geopolitical tensions may find common ground in their approach to SDG-relevant technologies, potentially opening avenues for collaboration in addressing global challenges.

The cluster analysis revealed five distinct groups of countries based on their technological research priorities related to Sustainable Development Goals. The “Eco-Tech Innovators” cluster consists of 20 countries: Australia, Bosnia and Herzegovina, Canada, Colombia, Croatia, Ecuador, Estonia, Greece, Indonesia, Kazakhstan, Kyrgyzstan, Lithuania, Malta, Namibia, Norway, Peru, Romania, Serbia, Turkey, and Ukraine. This group represents a diverse mix of nations focusing on environmental technologies.

The “Cyber-Digital Architects” cluster comprises 11 countries: China, Cyprus, Finland, Ireland, Liechtenstein, Luxembourg, Qatar, Singapore, South Korea, Taiwan, and the United Arab Emirates. These nations are known for their advanced digital infrastructures and focus on cybersecurity and digital technologies.

The “Bio-Industrial Pioneers” group is the second largest, with 31 countries: Austria, Belarus, Belgium, Brazil, Bulgaria, Chile, Costa Rica, Czechia, Denmark, France, Germany, Hungary, Israel, Italy, Japan, Latvia, Mexico, the Netherlands, New Zealand, Poland, Portugal, Russia, Slovakia, Slovenia, South Africa, Spain, Sweden, Switzerland, Thailand, the United Kingdom, and the United States. These countries are at the forefront of biotechnology and industrial innovation.

The “Geo-Data Security Analysts” cluster includes 22 countries: Argentina, Armenia, Bolivia, Botswana, Cameroon, Cuba, Iceland, Jamaica, Kenya, Libya, Moldova, Nepal, Panama, the Philippines, Rwanda, Syria, Tanzania, Trinidad and Tobago, Uganda, Uruguay, Venezuela, and Zambia. This group focuses on geographical information systems and data security.

Finally, the “Cyber-Sustainable Integrators” cluster, which is the largest, encompasses 36 countries: Afghanistan, Albania, Algeria, Azerbaijan, Bahrain, Bangladesh, Brunei, Egypt, Ethiopia, Fiji, Ghana, India, Iran, Iraq, Jordan, Kosovo, Kuwait, Lebanon, Malaysia, Mauritius, Montenegro, Morocco, Myanmar, Nigeria, North Macedonia, Oman, Pakistan, Palestine, Saudi Arabia, Senegal, Sri Lanka, Sudan, Tunisia, Uzbekistan, Vietnam, and Yemen. These nations are characterized by their efforts to integrate cybersecurity with sustainable development initiatives.

Cluster 1, the Eco-Tech Innovators, includes countries like Australia, Canada, and Norway, which are known for their significant natural resources and environmental challenges. This cluster’s focus on renewable energy and Robotics Automation aligns with these nations’ efforts to transition from resource-dependent economies to more sustainable models. For instance, Norway’s sovereign wealth fund, built on oil revenues, is increasingly investing in renewable technologies. Australia, facing severe climate change impacts like bushfires and coral reef degradation, has a growing impetus to innovate in environmental technologies. While most European countries are in Cluster 3, Estonia, known for its advanced digital society, is in Cluster 1. This could reflect Estonia’s particular focus on combining digital innovation with environmental sustainability, as seen in its “e-Estonia” initiative (Government of Estonia, 2024) [69].

Cluster 2, the Cyber-Digital Architects, features tech powerhouses like China, South Korea, and Singapore. These countries have made digital transformation a national priority. China’s “Made in China 2025” initiative (ISDP, 2018) [70] and Singapore’s “Smart Nation” program (Singapore Government Agency, 2024) [71] exemplify their commitment to leading in areas like the Internet of Things and next-generation networks. The presence of Finland in this cluster is interesting, possibly reflecting a strong focus on cybersecurity.

Cluster 3, the Bio-Industrial Pioneers, includes many of the world’s leading economies such as the United States, Germany, Japan, and the United Kingdom. These nations have traditionally been at the forefront of industrial innovation and have strong robotics and biotechnology sectors. The U.S., for example, leads in biotech research and development, while Germany’s “Industry 4.0” initiative (Schroeder, 2016) [72] focuses on advanced manufacturing. Japan’s strong robotics tradition and manufacturing base are enhanced by its “Society 5.0” initiative (Cabinet Office of Japan, 2024) [73], which aims to integrate cyber and physical spaces in advanced manufacturing and biotechnology. The inclusion of Russia in this cluster is noteworthy, possibly indicating its efforts to diversify from a resource-based economy and invest in high-tech industries, in the context of geopolitical tensions with other countries in the group.

Cluster 4, the Geo-Data Security Analysts, includes countries like Argentina, Kenya, and the Philippines. This grouping might reflect these nations’ focus on leveraging geospatial technologies and data security to address specific developmental challenges. For example, Kenya has been a leader in mobile banking and fintech in Africa, which requires robust data security. Argentina’s presence might relate to its efforts to modernize its agricultural sector, a key economic driver, using precision agriculture technologies.

Cluster 5, the Cyber-Sustainable Integrators, predominantly features countries from the Middle East, North Africa, and South Asia, including Saudi Arabia, Egypt, India, and Pakistan. This clustering could reflect these regions’ dual focus on rapid digital transformation and sustainable development. For instance, Saudi Arabia’s Vision 2030 (Kingdom of Saudi Arabia, 2024) [74] aims to diversify the economy away from oil, emphasizing both digital technologies and sustainability. India’s ambitious solar energy goals, coupled with its digital initiatives like “Digital India” (Government of India, 2024) [75], align well with this cluster’s priorities.

4. Discussion

This study presents an analysis of global research priorities for technologies relevant to Sustainable Development Goals (SDGs). By identifying and examining 18 different technological domains highly relevant for SDGs, it provides a broad view of the research landscape across countries. This study introduces an innovative within-country rank metric for each technology, which normalizes differences in overall research output between countries. This approach allows for meaningful comparisons of priorities even between nations with vastly different total productivity levels.

The K-means cluster analysis based on research priorities identifies five distinct clusters of countries with similar research focus patterns. This method goes beyond simple productivity metrics and uncovers unexpected research specializations, particularly in developing countries. The study’s dual approach, combining linear regression to predict overall research productivity with cluster analysis of research priorities, offers information about both the factors influencing overall productivity and the specific technological focus areas of different country groups.

The cluster analysis reveals some surprising groupings of countries based on their research priorities. For instance, Estonia is grouped with Australia and Canada in the “Eco-Tech Innovators” cluster, while China is grouped with Finland in the “Cyber-Digital Architects” cluster. These unexpected associations suggest that countries with historically different economic models or even geopolitical tensions may find common ground in their approach to SDG-relevant technologies.

The analysis identifies Artificial Intelligence as a consistently high priority across all clusters, while also revealing relatively underexplored areas such as Natural Language Processing and digital twin simulation. This information can guide future research investments and policy decisions. The geographical distribution of clusters reveals interesting regional patterns, such as the concentration of “Cyber-Sustainable Integrators” in the Middle East, North Africa, and South Asia.

Interestingly, the study reveals some discrepancies between the theoretical importance of certain technologies, as discussed in the literature, and their actual research prioritization. For example, blockchain technology and digital twins are emphasized in the literature but rank relatively low in research priorities across all clusters.

4.1. Cluster Analysis in Relation to Linear Regression Modeling

The K-means cluster analysis of countries’ research priorities among 18 SDG-relevant technologies offers a perspective that extends beyond the information provided by total research productivity metrics alone. This analysis shows distinct strategic approaches to addressing Sustainable Development Goals (SDGs) through technological innovation. For instance, the “Eco-Tech Innovators” cluster, which includes countries like Australia, Canada, and Norway, demonstrates a focus on renewable energy and robotics. This emphasis suggests these nations are prioritizing sustainable energy solutions and advanced automation to address environmental challenges. In contrast, the “Cyber-Digital Architects” cluster, comprising countries such as Singapore, South Korea, and Finland, places a higher priority on Internet of Things (IoT) and cybersecurity technologies. This focus indicates a strategy centered on building robust digital infrastructures to support sustainable development.

The cluster analysis also indicates trends and potential gaps in SDG-related technological development. Artificial Intelligence (AI) consistently ranks as a top priority across all clusters, underscoring its perceived importance in addressing various SDG challenges. However, the analysis also reveals technologies that are relatively underutilized, such as Natural Language Processing and digital twin simulation. This information can guide researchers and policymakers in identifying areas that may require increased attention and investment to fully leverage their potential for sustainable development.

Furthermore, the clustering reflects contextual factors that may not be apparent from total productivity figures. For example, the “Bio-Industrial Pioneers” cluster, which includes countries like Germany, Japan, and the United States, shows a strong focus on biotechnology and genetics alongside robotics and automation. This pattern may reflect these countries’ established industrial bases and their strategy to integrate advanced biological sciences with manufacturing technologies. This information can inform more targeted and effective policy decisions that align with a country’s existing strengths and economic structures.

The cluster analysis also facilitates the identification of collaboration opportunities. Countries within the same cluster, such as China and Ireland in the “Cyber-Digital Architects” group, may find natural partners for joint research initiatives or technology transfer programs due to their shared priorities in areas like IoT and cybersecurity. Conversely, countries from different clusters could engage in complementary collaborations, leveraging their diverse strengths. For instance, an “Eco-Tech Innovator” like Canada could partner with a “Bio-Industrial Pioneer” like the Netherlands to explore the intersection of renewable energy and biotechnology applications.

This clustering approach also provides valuable information for resource allocation. Organizations and governments can use this information to make more informed decisions about research funding and infrastructure development. For example, countries in the “Cyber-Sustainable Integrators” cluster, such as India and Malaysia, might choose to reinforce their strengths in renewable energy and IoT while also addressing potential gaps in areas like biotechnology or advanced manufacturing technologies. Lastly, the cluster analysis reveals comparative advantages that are not evident from total productivity metrics alone. For instance, the “Geo-Data Security Analysts” cluster, which includes countries like Argentina and Kenya, shows a focus on Geographic Information Systems combined with interests in biometric identity and Natural Language Processing. This specialized combination of research priorities could position these countries as valuable contributors to specific niche areas within the broader landscape of SDG-related technological innovation.

The cluster analysis also complements the regression model by pointing to differences in research priorities among countries with similar socio-economic indicators. While the regression model shows that GNI per capita and research funding are the strongest predictors of total research productivity, the cluster analysis demonstrates that countries with comparable levels of these indicators can have different research focuses. For instance, Canada and the Netherlands, both with high GNI per capita around USD 55,000–USD 60,000, fall into different clusters. Canada is classified as an “Eco-Tech Innovator”, likely reflecting its focus on environmental technologies driven by vast natural resources and environmental challenges. In contrast, the Netherlands is a “Bio-Industrial Pioneer”, possibly due to its strong agricultural and biotechnology sectors. Japan and Germany, with similar GNI per capita of approximately USD 45,000–USD 55,000, both fall into the “Bio-Industrial Pioneers” cluster. This shared focus on integrating advanced manufacturing with biotechnology could be attributed to their strong industrial bases and the challenges posed by aging populations, driving research in automation and healthcare technologies. Russia and Poland, with similar GNI per capita of about USD 15,000–USD 20,000, also fall into different clusters. Russia’s classification as a “Bio-Industrial Pioneer” might reflect its historical strength in industrial and biological sciences. Poland’s status as an “Eco-Tech Innovator” could be influenced by more recent emphases on renewable energy and environmental technologies, potentially driven by EU environmental policies. Brazil and China, despite having similar GNI per capita around USD 9000–USD 13,000, show distinct research focuses. Brazil’s categorization as a “Bio-Industrial Pioneer” likely stems from its emphasis on agricultural and biofuel technologies, leveraging its natural resources. China’s classification as a “Cyber-Digital Architect” country aligns with its strategic push in digital technologies and AI. These differences in research priorities among economically similar countries can be explained by various factors, including historical industrial strengths, geographic and environmental considerations, strategic national policies, regional influences, and specific societal challenges.

The cluster analysis also highlights potential matches or mismatches between research priorities and development needs. For example, Indonesia, classified as an “Eco-Tech Innovator”, shows a focus on renewable energy research. This aligns well with its archipelagic geography and vulnerability to climate change, demonstrating a strategic match between research priorities and national challenges.

Furthermore, the cluster analysis can help identify outliers or unexpected patterns, revealing cases that are not apparent from the regression model alone. For instance, some developing countries with lower GNI per capita and research funding fall into clusters typically associated with more advanced economies, indicating targeted investments in specific technological areas or successful niche strategies. A notable example is India, which, despite its lower GNI per capita compared to many advanced economies, is classified as a “Cyber-Sustainable Integrator”. This classification indicates a focus on integrating digital technologies with sustainable development goals, particularly in areas like renewable energy and the Internet of Things. India’s position in this cluster, alongside countries with higher GNI per capita like Bahrain and Kuwait, indicates a strategic emphasis on digital and sustainable technologies that goes beyond what might be expected based solely on economic indicators. Kenya provides another example of a lower-income country showing research priorities shared with countries with higher values for these indicators. Classified as a “Geo-Data Security Analyst”, Kenya shares a cluster with countries like Iceland and Uruguay. This shows a focus on Geographic Information Systems and data security technologies, possibly driven by Kenya’s emerging role as a technology hub in East Africa and its efforts to address environmental and security challenges through technological solutions.

4.2. Linear Regression Modeling in Relation to the State of the Art

The findings from the regression analysis largely align with the state-of-the-art literature on factors influencing scientific productivity, particularly in the context of technologies supporting Sustainable Development Goals (SDGs). The regression model identifies economic factors, specifically GNI per capita and research spending as a proportion of GDP, as the strongest predictors of a country’s research productivity in SDG technologies. This is consistent with several studies on the state of the art that emphasize the importance of economic wealth and R&D expenditure in driving scientific output (Rodríguez-Navarro and Brito, 2023; Lancho-Barrantes et al., 2023; Rahman and Fukui, 2022) [52,54,57]. The strong positive relationship between economic indicators and research productivity supports the notion that countries with greater economic resources can invest more in research infrastructure and human capital, leading to higher scientific output.

The regression analysis also found a positive relationship between life expectancy and research productivity. While this specific indicator is not directly mentioned in the state of the art, it could be interpreted as a proxy for overall development and quality of life. This aligns with findings that suggest societal factors play a role in shaping scientific output (Allik et al., 2023) [60].

The positive association between the liberal democracy index and research productivity in the regression model is consistent with the literature that highlights the importance of good governance and institutional quality in fostering scientific productivity (Lancho-Barrantes et al., 2023; Allik et al., 2023) [54,60]. This supports the idea that democratic societies may provide better conditions for research, possibly through academic freedom and support for scientific institutions.

However, the regression analysis diverges from some aspects of the state of the art. For instance, the model found no significant relationship between average years of schooling and research productivity, when controlling for the other indicators, which contrasts with research emphasizing the role of educational infrastructure in scientific output (Jamjoom and Jamjoom, 2022) [56]. This discrepancy might be due to the specific focus on SDG technologies in the regression model, or it could suggest that other factors overshadow the direct impact of education when considering multiple variables simultaneously.

The regression model also did not find a significant relationship between the SDG score and research productivity. This is an interesting divergence from what might be expected based on the state of the art, which suggests that environmental policies can predict research output in related fields (Dragos and Dragos, 2022) [61]. This discrepancy could indicate that a country’s overall SDG performance may not directly translate into increased research productivity in SDG-related technologies, when controlling for other socio-economic indicators.

4.3. Cluster Analysis in Relation to the State of the Art

The findings from the cluster analysis largely corroborate the state of the art in SDG-relevant technologies, while also revealing some divergences. Both emphasize the critical role of Artificial Intelligence in advancing SDGs. The state of the art highlights AI’s diverse applications, from enhancing healthcare delivery to optimizing urban planning and improving decision-making through effective data management (ITU and UNDP, 2023; Adel and Alani, 2024; Bachman et al., 2022) [2,6,7]. This importance is reflected in the cluster analysis, where AI consistently ranks as the top research priority across all country clusters.

Renewable energy technologies are similarly emphasized in both the state of the art and the cluster analysis. The literature review identifies these technologies as fundamental in addressing climate change and promoting sustainable energy access (ITU and UNDP, 2023; United Nations, 2023; World Economic Forum, 2020) [2,3,4]. This aligns closely with the cluster analysis findings, which show renewable energy as a high priority, particularly for the “Eco-Tech Innovators” and “Cyber-Sustainable Integrators” clusters. This alignment suggests a global recognition of the critical role renewable energy plays in achieving sustainability goals, as also noted by Bachman et al. (2022) [7].

Cybersecurity emerges as another area of concordance between the state of the art and the cluster analysis. The literature review stresses the importance of cybersecurity in maintaining trust in digital systems and safeguarding critical infrastructure (UNDP, 2023; ITU and UNDP, 2023; United Nations, 2023) [1,2,3]. This emphasis is mirrored in the cluster analysis, which reveals cybersecurity as a key focus for the “Cyber-Digital Architects” and “Cyber-Sustainable Integrators” clusters, indicating a strategic prioritization of digital security in certain country groups. This aligns with the observations of Palomares et al. (2021) [21] and Michael et al. (2019) [22] on the critical role of cybersecurity in enabling other Industry 4.0 technologies.

The Internet of Things (IoT) and Big Data Analytics are highlighted in the state of the art as essential for resource management and decision-making (World Economic Forum, 2020; Kasinathan et al., 2022; Wu et al., 2018) [4,17,20]. The cluster analysis reflects this importance, showing IoT as a high priority for some clusters and Big Data Analytics maintaining moderate importance across all clusters. This shows a widespread recognition of the value these technologies bring to sustainable development efforts, as also noted by Hassoun et al. (2022) [18] and Adel and Alani (2024) [6].

Robotics and automation are another area where the state of the art and cluster analysis findings align. Both sources highlight the role of these technologies in healthcare, agriculture, and sustainable production (Kasinathan et al., 2022; Hassoun et al., 2022; Mabkhot et al., 2021) [17,18,24]. The cluster analysis further refines this understanding by showing robotics and automation as top priorities specifically for the “Bio-Industrial Pioneers” cluster, suggesting a concentration of research efforts in certain countries, which aligns with observations by Bachman et al. (2022) [7] on the role of these technologies in sustainable production methods.

However, the cluster analysis also reveals some divergences from the state of the art. While Geographic Information Systems (GISs) are mentioned in the literature as critical for various SDG applications (UNDP, 2023; Bachman et al., 2022) [1,7], the cluster analysis shows them as a top priority only for the “Geo-Data Security Analysts” cluster. This suggests a more specialized focus on GISs than might be expected based on their theoretical importance, as discussed by Wu et al. (2018) [20] in the context of environmental SDGs. Similarly, blockchain technology is emphasized in the state of the art for its potential in enhancing transparency and traceability in supply chains and governance processes (ITU and UNDP, 2023; Parmentola et al., 2022; De Villiers, Kuruppu, and Dissanayake, 2021) [2,33,41]. However, the cluster analysis indicates that blockchain is a relatively low priority across all clusters, pointing to a possible gap between its theoretical potential and actual research focus, a discrepancy also noted by Adams, Kewell, and Parry (2018) [43] in their discussion of blockchain’s environmental impacts.

Lastly, while digital twins and simulation technologies are mentioned in the state of the art as enablers of better planning and management of complex systems (World Economic Forum, 2020; Palomares et al., 2021; Mabkhot et al., 2021) [4,21,24], they rank low in research priorities across all clusters in the analysis. This discrepancy might indicate a lag between recognizing the potential of these technologies and allocating significant research resources to them, a challenge also identified by Goh and Vinuesa (2021) [40] in their discussion of AI implementation for SDGs.

4.4. Research Limitations

This study’s findings should be interpreted in light of several methodological limitations. Primarily, the reliance on Web of Science publication data underrepresents research from non-English-speaking countries or those with less established research infrastructures. The classification of research into technological domains based on keyword queries introduces potential imprecision, possibly missing nuanced or interdisciplinary work that does not use standard terminology. The focus on published research excludes ongoing studies, works in progress, or research disseminated through non-traditional channels.

It is important to note that this study measures research productivity primarily in terms of publication count, which does not account for the quality, impact, or practical application of the research. The analysis is also limited by the socio-economic indicators included, potentially overlooking other influential factors such as international collaborations or specific national policies.

Our study provides a cross-sectional analysis of global research priorities at a single point in time, which limits our ability to capture temporal changes in research focus. This approach does not account for evolving global challenges or technological advancements that could shift priorities over time. Future studies could benefit from incorporating longitudinal data to track changes in research priorities across multiple time points.

Another limitation of this study is the lack of consideration for the state of the entrepreneurial ecosystem (also known as the innovation ecosystem) in each country. This ecosystem shapes the level of commercialization of scientific developments and their introduction into national economies, influencing economic growth. The entrepreneurial ecosystem affects scientific productivity by providing pathways for research to be translated into practical applications and by creating feedback loops between industry needs and academic research priorities. Our analysis does not account for these dynamics. Future research could benefit from incorporating measures of the entrepreneurial ecosystem, such as those provided by international indices and rankings, to gain a better understanding of the factors influencing technological innovation in support of Sustainable Development Goals.

Our study employs K-means clustering to categorize countries based on their research priorities in SDG-relevant technologies. While this method is widely used and offers valuable information, it has inherent limitations. One significant constraint is the requirement to pre-specify the number of clusters, which may not fully capture the typologies of research priorities across diverse countries. This predetermination could lead to oversimplification or artificial groupings that may not accurately reflect the full spectrum of global research focus patterns. The fixed number of clusters might force countries with somewhat different priorities into the same group or separate countries with similar profiles. Alternative clustering methods, such as hierarchical clustering or density-based clustering, might reveal different patterns or groupings. Future research could benefit from employing multiple clustering techniques or more advanced methods that do not require pre-specifying the number of clusters. These approaches could provide a more flexible and potentially more accurate representation of the diverse landscape of global research priorities in SDG-relevant technologies. Furthermore, incorporating qualitative analyses or expert validation of the clusters could enhance the robustness and interpretability of the results. Furthermore, the national-level analysis may obscure important sub-national variations, particularly in larger or more diverse countries.

While comprehensive, the 18 technological domains selected may not exhaustively cover all technologies relevant to SDGs, potentially underrepresenting emerging or niche technologies. Lastly, the cross-sectional nature of the study provides a snapshot of research priorities at a single point in time, without capturing longitudinal trends or shifts in research focus.

4.5. Future Research

Building upon the current study’s findings and addressing its limitations, future research in this field could explore several promising directions. A key area for expansion is the diversification of data sources. Incorporating research outputs from a broader range of databases, including regional and non-English-language repositories, could provide a more comprehensive global picture of SDG-relevant technological research. This approach would help mitigate the potential underrepresentation of research from non-Western or developing countries.

To enhance the precision of technological domain classification, future studies could employ more sophisticated text analysis techniques, such as Natural Language Processing and machine learning algorithms. These methods could better capture interdisciplinary research, providing a more accurate representation of the global research landscape. Additionally, integrating alternative metrics of research impact, such as citation counts, patents, or practical applications of research findings, would offer a broader view of a country’s contribution to SDG-related technologies beyond publication counts.

Longitudinal studies tracking research priorities over time would be highly valuable. Such analyses could reveal how national and global research focuses evolve in response to changing technological landscapes, policy environments, and societal needs. This temporal dimension would provide information about the responsiveness of the research community to emerging challenges and opportunities in sustainable development.

Expanding the scope of socio-economic indicators in future research could yield richer insights into the factors driving research priorities. Including variables such as international collaboration networks, specific national science policies, private-sector R&D investments, and measures of institutional quality could provide a better understanding of the determinants of research focus and productivity.

To address the potential underrepresentation of emerging technologies, future research could employ more flexible and adaptive frameworks for identifying and categorizing technological domains. This might involve regular consultations with experts across various fields to ensure the inclusion of cutting-edge and niche technologies relevant to SDGs. Finally, complementing quantitative analyses with qualitative research methods, such as case studies of specific countries or research institutions, or interviews with policymakers and researchers, could provide deeper context and explanations for the patterns observed in the data.