ADCAIJ: Advances in Distributed Computing and Artificial Intelligence Journal
Regular Issue, Vol. 14 (2025), e32890
eISSN: 2255-2863
DOI: https://doi.org/10.14201/adcaij.32890

AI for Solving SDGs: A Trend Analysis of Public Interest in Sustainable Development Goals and Artificial Intelligence in Africa

Kabir Opeyemi Olorede^a, and Abejide Ade-Ibijola^b

^aDepartment of Statistics and Mathematical Sciences, Faculty of Pure and Applied Sciences, P.M.B. 1530, Kwara State University, Malete, Kwara State, Nigeria

^bAI Research Group, GRIT Lab Africa, South Africa

kabir.olorede@kwasu.edu.ng, abejide@gritlabafrica.org

ABSTRACT

KEYWORDS

1. Introduction

The 17 United Nations Sustainable Development Goals (SDGs) present a global framework for addressing pressing challenges, ranging from poverty and hunger through industrial innovation and climate change to health and partnership for the goals. As these goals seek a more sustainable and equitable future, the role of emerging technologies such as artificial intelligence (AI) has become increasingly prominent (Singh et al., 2024). AI holds the potential to accelerate progress towards achieving the SDGs by offering innovative solutions across diverse sectors, including business, healthcare, agriculture, education, and energy. At the same time, AI presents both opportunities and challenges for African nations that face distinctive socio-economic and infrastructural constraints.

Africa’s heterogeneous development landscape, characterised by uneven technological infrastructure, varying levels of digital literacy, and substantial economic disparities, provides a critical context for examining how AI can support the achievement of the SDGs. While some African countries have made considerable progress in developing policies and strategies that enable AI adoption, others remain constrained by limited resources, institutional capacity, and access to digital infrastructure. Understanding how AI adoption interests align with SDG priorities across African countries is therefore essential for promoting equitable development and realising AI’s potential for social good. Despite this importance, there remains limited empirical research that examines the convergence of AI innovation adoption and SDG related interests across African regions using real time data sources.

This study addresses that gap by analysing public search interest in AI and the SDGs across Africa using Google Trends Explore (GTE) data (Google Trends, 2024). Search behaviour is treated as a behavioural proxy for public awareness, engagement, and emerging readiness to apply AI to sustainable development challenges. To capture intermittent and sparse series while revealing underlying structure, we employ Croston’s intermittent time series trend estimation, agglomerative hierarchical clustering on principal components, and visual analytics (Croston, 1972; Svetunkov & Boylan, 2023; Husson et al., 2010; Lê et al., 2008). The dataset covers monthly GTE search interest for AI and SDG related topics across African countries over nine years from September 2015 to August 2024.

Within this framework, the inquiry is explicitly guided by three research questions embedded in the analysis: first, what are the current relative prominence indices of African regions when AI is considered for solving the SDGs; second, which African sub regions presently rank higher in adoption readiness for solving the SDGs with AI; and third, in what ways do regional AI and SDG engagement trends differ across the continent over time. These questions are operationalised through the computation of relative salience indices, trend estimation on intermittent search activity, and clustering to uncover latent regional structures. The resulting evidence offers a comparative and empirically grounded view of where interest in leveraging AI for the SDGs is strongest, where readiness is emerging, and where gaps persist, thereby informing decision making by policymakers and relevant stakeholders.

1.1. Structure and Flow of the Paper

The remainder of this paper is organised as follows. Section 2 reviews the relevant global and Africa-focused literature on AI and the sustainable development goals. Section 3 describes the data sources and analytical techniques, including the use of Google Trends Explore data, intermittent time series forecasting, and hierarchical clustering. Section 4 presents and discusses the empirical results, with region-specific analyses of AI and SDG search trends across Africa. Section 5 synthesises regional patterns and comparative findings. Section 6 outlines key challenges and limitations of the study. Finally, Section 7 concludes the paper by summarising the main findings, discussing policy implications, and identifying directions for future research.

2. AI for Solving SDGs Globally

As a transformative tool in addressing many challenges outlined by the United Nations’ Sustainable Development Goals (SDGs), the literature on artificial intelligence (AI) has grown significantly. This reflects the expanding role of technological innovation in addressing global challenges across various domains. Globally, AI has been applied in diverse ways to tackle poverty, healthcare, education, and environmental sustainability issues. Goralski and Tan (2020) explore the disruptive impact of AI on sustainable development, emphasising the potential for AI to transform industries and support sustainable practices across sectors. However, their analysis identifies a limited focus on educational implications and the integration of AI in management training, underscoring a need for broader sectoral integration of AI to realise its full potential across SDGs.

Vinuesa et al. (2020) expand on the relevance of AI for solving SDGs, particularly its applications in monitoring and evaluating progress in areas like poverty reduction, health, and education. While the study highlights AI’s potential for accelerating SDG advancements, it cautions about the ethical and technical considerations, advocating for responsible usage to avoid data privacy and security issues. Hoosain et al. (2020) examine the impact of Fourth Industrial Revolution (4IR) technologies, including AI, on the SDGs. They discuss the transformative potential of digital tools, suggesting that AI and circular thinking can foster sustainable development in various sectors. This article encourages further research to understand how these technologies can integrate and support SDGs, especially in contexts that lack robust infrastructure for widespread AI adoption.

In 2021, Astobiza et al. (2021) focus specifically on the ethical dimensions of AI in achieving SDGs. Their work stresses the importance of designing ethical frameworks that prevent biases and ensure AI applications are aligned with sustainable development needs. Similarly, Holzinger et al. (2021) discuss AI’s transformative role in SDG digital transformation, focusing on security, safety, and privacy challenges. They highlight the need for transparency, trustworthiness, and global standards to guide AI in sensitive areas such as health and agriculture. Arfanuzzaman (2021) examines the potential of AI and big data in promoting sustainable urbanisation in South Asia, highlighting challenges in infrastructure and partnerships that hinder AI’s leverage for SDGs. Although focused on South Asia, the study underscores the critical role of AI and big data in solving sustainable urbanisation, which is relevant to global urban contexts. Roy et al. (2022) focus on Sustainable Development Goal 6 (SDG6) related to water and sanitation, using scientometric and text mining techniques to analyse research trends and collaboration patterns. Their study reveals a lack of engagement from developing nations in SDG 6 research, proposing a framework to guide future AI-driven research efforts to enhance data analysis in solving water and sanitation problems.

Ramezani et al. (2023) provide a comprehensive bibliometric analysis of AI trends in solving health-related SDGs, showing a sharp rise in research from 2015-2022. They highlight a global perspective on health and AI innovation, emphasising the need for governance to support sustainable health outcomes. Similarly, Fazal et al. (2024) maps COVID-19-related research linked to solving SDGs, pinpointing the importance of resilience strategies and sustainable decision-making in a post-pandemic world. Singh et al. (2024) analyse bibliometric trends in AI applications for SDGs, identifying regional disparities and dichotomies in AI research participation. They encourage investment in areas where AI can significantly and progressively solve SDGs, addressing existing research gaps that limit a comprehensive understanding of AI’s regional impact.

Finally, Fazal et al. (2024) discuss AI’s role in promoting financial inclusion, a critical enabler for solving multiple SDGs. Their work highlights AI’s applications in financial services, such as M-Pesa, showing how AI-driven financial inclusion can support economic growth and poverty reduction. However, the study is limited by its focus on a small number of AI applications. Together, these studies reveal a trajectory where AI’s application to SDGs is increasingly multi-faceted but still requires ethical, technical, and infrastructural improvements to maximise its potential across regions and sectors. The literature highlights the need for more region-specific studies, especially in developing countries, alongside ethical guidelines and robust indicators to track AI’s effectiveness in achieving SDGs. This evolving body of work underscores AI’s potential for transformative impact while highlighting essential considerations for ensuring its responsible deployment in sustainable development initiatives.

3. Africa’s Technological Landscape

Africa’s technological landscape is evolving rapidly, driven by digital transformation initiatives and increased mobile and internet penetration. In recent years, Africa has seen a rise in tech hubs, startup ecosystems, and innovation centres, particularly in the Republic of South Africa, the Federal Republic of Nigeria, the Republic of Kenya, and the Arab Republic of Egypt. These developments have catalysed the use of AI in sectors such as agriculture, healthcare, and financial services.

Despite these advances, the adoption of AI in Africa remains uneven. Limited infrastructure, insufficient digital literacy, and uneven access to quality data have slowed AI development in many regions. The continent’s digital divide, particularly between urban and rural areas, hinders the full realisation of AI’s potential in solving the SDGs. Nonetheless, initiatives such as the African Union’s Digital Transformation Strategy for Africa (African Union, 2020) demonstrate a growing recognition of AI’s importance in driving sustainable development on the continent. AI’s application to solving the SDGs has grown significantly, mainly as frameworks, models, and innovations focus on global and Africa-specific development challenges. Steingard et al. (2023) recently introduced an SDG-Intense Evaluation framework to align academic AI applications with SDGs, underscoring AI’s transformative potential in academia and its indirect impact on African development. However, it lacks Africa-specific implementation insights. Similarly, Mukonza and Chiang (2023) demonstrate the utility of machine learning in satellite-based water quality monitoring, which holds relevance for SDG 6 on clean water. However, it emphasises non-African contexts, overlooking Africa’s unique water management needs.

Sinde et al. (2023) provide a crucial framework for responsible AI adoption in Anglophone Africa, urging partnerships between academia and the private sector to strengthen ethical, infrastructure-ready AI integration essential to addressing Africa’s development needs. In their discussion on Green Supply Chain Management (GSCM) and SDGs, Raman et al. (2023) note that AI-driven GSCM practices, while promising for environmental SDGs, lack regional focus, highlighting an opportunity for African supply chains to adopt sustainable practices amid rising interest. Addressing regional governance challenges, Ade-Ibijola and Okonkwo (2023) emphasise the need for effective governance to facilitate AI’s productive potential across Africa.

Ajaj et al. (2024) explore Industry 4.0 to emphasise AI and other advanced technologies in solving poverty (SDG1), which is relevant to African contexts though lacking concrete case studies. Similarly, Arthur et al. (2024) review the integration of AI and blockchain for tackling financial crimes in Africa, which could strengthen institutions, align with SDG 16, and support economic growth. Wakunuma et al. (2022) provide an in-depth examination of responsible AI governance to meet SDGs in Africa, particularly SDG 3 on health, identifying the absence of regulatory frameworks as a significant barrier to AI adoption.

In FinTech, Hoang et al. (2022) discuss how innovations in digital finance, such as AI and blockchain, are essential for financial inclusion and SDGs but face financial and infrastructural challenges specific to Africa. Meanwhile, Sampene et al. (2022) explore AI’s role in economic transformation in Africa, noting that while AI holds significant promise for sectors like healthcare and agriculture, infrastructural and ethical concerns hinder its potential. Additionally, Bjola (2022) reviews AI’s role in rethinking development approaches, advocating for an ethical integration of AI in policy and practice to enhance SDGs, though without an African focus.

Earlier studies, such as Mhlanga (2020) examine AI in solving poverty reduction needs within the Fourth Industrial Revolution (4IR), emphasising the need for advanced AI tools like poverty mapping to enhance resource allocation in Africa. Salvia et al. (2019) analyse regional SDG trends, identifying Africa’s priorities as poverty and sanitation (SDGs 1 and 6), which informs the need for tailored AI solutions. Lastly, Pedro et al. (2019) highlight AI’s role in improving educational outcomes (SDG4), advocating for frameworks that could be transformative for Africa’s educational challenges despite limited contextual focus.

These works highlight AI’s transformative potential across multiple SDGs in Africa yet underscore the continent’s unique barriers. Key challenges such as limited infrastructure, regulatory gaps, and the need for ethical frameworks emerge as critical areas requiring targeted research and collaborative governance to fully realise AI’s promise in advancing Africa’s development agenda. Notably, these studies completely ignored the need for an empirical investigation of the regional trends and ranking of the African countries in terms of their readiness and current strengths in adopting AI for solving the SDGs. This is essential to simplify Africa’s currently complex adoption and implementation patterns.

4. AI in Solving SDGs in Developing and Emerging Regions

AI has shown promise in developing regions, although with some obvious limitations. Projects in South Asia and Latin America have demonstrated AI’s ability to solve and improve education outcomes, provide financial inclusion through fintech solutions, and support small-scale businesses (Chui et al., 2018). However, the success of AI in solving the SDGs varies significantly, depending mainly on the availability of technology, data, and policy frameworks that encourage innovation and equitable use of AI tools. Truby (2020) raises critical concerns about unregulated AI deployment by large technology corporations, particularly its potential to undermine SDG progress in developing countries. The study further emphasises the need for strong regulatory frameworks to ensure AI initiatives align with SDG priorities. He argues for AI governance models that include regular audits and compliance measures, noting the absence of a robust, internationally coordinated regulatory framework. This work underscores that governance structures must be in place to support sustainable and equitable AI growth, especially where unregulated deployment risks worsening existing inequalities.

Mhlanga (2021) emphasises AI’s potential to reduce poverty and drive innovation in emerging economies. This research highlights AI’s transformative role in enhancing infrastructure and fostering economic growth but points out that its impact varies significantly across economies. Mhlanga calls for increased government support to maximise AI’s poverty reduction and development potential. However, he also identifies a gap in longitudinal studies that would more effectively capture AI’s sustained impact across different economic contexts. This suggests a need for more comprehensive, comparative analyses to deepen our understanding of AI’s influence on SDGs in emerging economies. Most recently, Meitei et al. (2023) provide a comprehensive overview of trends in AI and ML applications targeting how to solve SDGs over a five-year period. This review focuses predominantly on contributions from developed countries, leaving a gap in understanding AI adoption in Africa or other emerging regions. The authors underline the need for ethical scrutiny and regulatory frameworks to guide AI applications toward sustainable outcomes, stressing responsible innovation and high-quality data as prerequisites for AI’s effective use in progressively solving SDGs. This work advocates responsible, well-regulated AI development, recognising its potential while cautioning against risks associated with its application.

These studies collectively highlight the potential and complexities associated with AI in the context of sustainable development. From the foundational call for governance frameworks (Truby, 2020) to the emphasis on tailored support and longitudinal studies for emerging economies (Mhlanga, 2021) and, finally, the critical need for responsible innovation backed by ethical oversight (Meitei et al., 2023), the literature underscores that the path to leveraging AI for SDGs must be conscientious, inclusive, and strategically regulated. There is also a complete omission of the research perspective explored in this study.

5. Limitations of Prior Work

While the global application of AI for achieving the SDGs has been widely studied, research specifically focused on Africa remains limited. Much of the existing literature highlights AI’s potential but lacks region-specific data and in-depth analysis of how AI is deployed and sought after in African countries. Moreover, few studies examine the alignment of AI with specific SDGs within the African context, and even fewer utilise real-time data to track trends and engagement with AI technologies. A notable literature gap is the absence of studies leveraging real-time search data, such as Google Trends, to gauge public interest and engagement with AI in relation to the SDGs. This analysis could provide valuable insights into how AI and SDG-related engagements are prioritised in different African sub-regions and how AI adoption evolves over time. By addressing this gap, this study aims to contribute a more dynamic and data-driven perspective to the growing discourse on AI and sustainable development in Africa.

6. Data and Analysis Techniques

6.1. Data Source

This study employs Google Trends Explore (Choi & Varian, 2012; Franzén, 2023; Google Trends, 2024; Jun et al., 2018; Mavragani et al., 2018), a web-based tool that provides insights into the frequency of specific search terms over a given period for analytic insights. Google Trends Explore (GTE) measures «search interest» on a scale from 0 to 100, where 100 indicates the peak search intensity of a search term and 0 indicates insufficient search data. The tool allows for comparing search interests across regions and categories, making it an effective method for identifying patterns and trends in online activity.

GTE provides indexed and normalised data, making it a practical proxy for public interest in AI and SDGs. However, this abstraction hides the true magnitude and distribution of searches, as GTE reports relative popularity rather than exact counts (Deloitte, 2025; Franzén, 2023; Google Cloud, 2025). The automatic exclusion of low-volume queries, omission of repeated searches, and removal of special characters can distort patterns, especially for small businesses or regions with sparse activity (Google Research Brief, 2024; Jun et al., 2018). These limitations raise concerns about reliability and applicability, yet GTE remains useful for tracking relative engagement trends, especially when complemented with statistical procedure that can handle the snags.

6.2. Data Collection

Data collection involved querying Google Trends with the search terms «Artificial Intelligence» and «Sustainable Development Goals» across the African Union member countries (African Union, 2024b) over nine years, from September 2015 to August 2024. The data were gathered monthly, ensuring that short-term fluctuations and long-term trends were captured. AI and SDGs constituted variables measuring search interests for AI-related concepts and the 17 SDGs. The regions covered include countries from all five African regions: Northern, Western, Eastern, Central, and Southern Africa. Each query provided a search interest score, which was used to measure public interest in AI and SDG-related topics.

6.3. Justification for Coverage

For this study, the search covers terms related to the United Nations’ Sustainable Development Goals (SDGs) and Artificial Intelligence (AI). The 2015–2024 timeframe for analysing the Google Trends search data on AI and SDGs is carefully chosen to capture critical developments and shifts in the relationship between AI and sustainable development within Africa. The adoption of the SDGs in 2015 marked a global push toward ambitious objectives, emphasising the potential of emerging technologies like AI to accelerate progress in healthcare, education, poverty reduction, and environmental protection (United Nations, 2015). This post-2015 period reflects an intensification in AI initiatives to address challenges aligned with these global goals, making it a pivotal starting point for the analysis.

Since 2017, Africa has seen a notable rise in AI-driven innovations tailored to regional needs, especially within healthcare, agriculture, and financial inclusion (Meitei et al., 2023). During this time, African nations introduced policies and frameworks, such as South Africa’s National AI Strategy and Kenya’s AI Policy Framework, which underscore responsible and sustainable AI development (Mhlanga, 2021; UNESCO, 2022, 2024). These national strategies emphasise using AI to promote regional growth and reduce inequalities, sparking widespread interest and engagement across various sectors. Google search interest captured during this period likely reflects this growing focus as governments, organisations, and the public seek insights into applying AI to advance sustainable development. The COVID-19 pandemic (Keesara et al., 2020; Roberts et al., 2021), spanning 2020 to 2022, further accelerated AI adoption in essential services, especially in public health contexts. This period saw intensified global and regional interest in AI’s potential to manage public health crises, as evidenced by AI applications for predictive modelling, remote diagnostics, and resource management (Truby, 2020). Public and institutional focus on health-related AI innovations increased interest, aligning with SDG 3 (Good Health and Well-Being) as stakeholders sought AI solutions for pressing health challenges.

Finally, the period from 2023 onward reflects an increased emphasis on ethical and regulatory considerations around AI, following the rapid expansion of these technologies. This period marks the African Union’s 2021 Continental Framework for AI, which underscores the need for responsible AI practices that protect individual rights and address social challenges (African Union, 2024a). Consequently, recent Google Trends data likely captures rising public awareness and institutional focus on AI ethics and regulation. By capturing public and institutional engagement with AI over nearly a decade, the selected period from 2015 to 2024 offers a robust dataset to assess how interest in AI and SDG alignment has evolved in response to key events, policies, and innovations. This time frame provides a comprehensive basis for analysing the broader trends in AI’s role in sustainable development, making it ideal for examining AI’s impact on the achievements of the SDGs.

6.4. Data Analysis Techniques

This study combines quantitative analysis and visualisation techniques to excavate the AI and SDGs search interest data sets to discover the hidden trends and insights on their salience and interconnectivity on the African continent. The GTE data utilise the relative salience of AI and SDG over time as normalised and indexed values instead of the exact count of the searches by automatically excluding and categorising search terms with low volume as zero for the relevant periods. This presents the GTE search interests data as sparse time series data due to non-zero values appearing sporadically in time, while others remain 0. Typical examples of sparse time series data include intermittent demands, such as demands for aviation spare parts (Shafi et al., 2023) and rainfall over time. We controlled for the effects of random (sampling) zeros due to the exclusion of search terms with low volumes for the relevant periods (Google Trends, 2024) and potential bias due to the indexed and normalised nature of the GTE data (Franzén, 2023) by smoothing the intermittent AI and SDG search interests, enabling more accurate and reliable trend estimation in periods with zero search activity using Croston’s method for intermittent demands (Croston, 1972; Svetunkov & Boylan, 2023). Croston’s method is particularly suitable for forecasting intermittent demand data, where many periods may have zero values, followed by occasional non-zero values. We know existing variants of Croston’s and other related methods. However, our choice of Croston’s method is due to its computational speed and ease of implementation in our computational algorithm for the study (Grekov et al., 2024).

The Croston method is often less effective after extended periods of zero demand because it updates forecasts only during periods of buoyant demand (Teunter et al., 2011). While this limitation poses challenges in inventory management, such as addressing product obsolescence or excess stock, it does not impact on our study on AI and SDG search trends. We focus on addressing sparseness by analysing irregular patterns in GTE data rather than managing inventory, making the method suitable for our research context. As a modification of the simple exponential smoothing (SES) method designed initially for modelling discontinuous time series, Croston’s method is described by the equation:

$$y_t={\omega }_t{z}_t (1)$$

where y_t is the actual search observation in the context of this study, ω_t is a Bernoulli-type variable that assumes a value of one when a detectable search occurs and zero otherwise, z_t denotes the potential search quantity, which becomes real only in moments when ω_t = 1, and t is the observation time. To achieve the ease of estimating the probability of a search occurring using intervals between searches, we considered δ_j search intervals and z_j search volumes, beyond single t = 1, ⋯, T series where j_t = 1, ⋯, N denotes the successive numbers of intervals and search volumes over N number of non-zero searches (Svetunkov & Boylan, 2023). Any quantity δ_jt representing the time since the last detectable (non-zero) search and reflecting the search interval then indicates the next detectable search volume. Then, forecasting both search volumes z_jt and search intervals δ_jt using SES (Grekov et al., 2024; Prestwich et al., 2021) can be achieved with the following system of equations:

$$\begin{array}{rl}& {\hat{y}}_t={\hat{y}}_{jt}=\frac{1}{{\delta }_{jt}}{\hat{z}}_{jt}\\ & {\hat{z}}_{jt}={\phi }_z{z}_{jt-1}+(1-{\phi }_z){\hat{z}}_{jt-1}\\ & {\hat{\delta }}_{jt}={\phi }_q{\delta }_{jt-1}+(1-{\phi }_q){\hat{\delta }}_{jt-1}\\ & j_t=j_{t-1}+{\omega }_t\end{array} (2)$$

where ${\hat{y}}_{jt}$ is the projected average search, ${\hat{Z}}_{jt}$ is the projected search volume, ${\hat{\delta }}_{jt}$ is the forecast search interval, and φ_q and φ_z are the smoothing options for search intervals and volumes (Kourentzes, 2014; Schultz, 1987; Snyder, 2002), respectively.

Using the system of Equations (2), we applied Croston’s method separately on each time series for SDG and AI search interests, helping to fill in gaps and smooth the data to better observe underlying trends in cases where there may be prolonged periods with little to no search activity (zero values). The method produces smoothed forecast search values for SDG and AI for each country. In line with the study objective of investigating AI for addressing the SDGs, these smoothed values were then used to calculate the Relative Salience Index (RSI):

$$RSI\mathit{ }\mathit{=}\frac{\mathit{SDG}\mathit{ }\mathit{Search}\mathit{ }\mathit{Volume}}{\mathit{(}\mathit{SDG}\mathit{ }\mathit{+}\mathit{ }\mathit{AI}\mathit{)}\mathit{ }\mathit{Search}\mathit{ }\mathit{Volume}} (3)$$

such that for any 0 ≤ RSI ≤ 1. The RSI measures SDG’s relative prevalence (interest) compared to AI over time for each country. RSI values close to 1 indicate that the SDG-related interest or activity is much higher than AI within the country for that period and vice versa. Values close to 0 suggest that AI-related interest or activity dominates SDG. Values around 0.5 imply a balanced interest or prominence of SDG and AI and vice versa. This estimation strategy provides a more consistent measure of interest between SDG and AI search intensity over time. Furthermore, we repeated the forecasts to align with the length of the original series, ensuring that they cover the entire timeline without introducing data length mismatch.

Time-series trend analysis alone cannot uncover latent groupings within search data. To address this, we applied Croston’s method, which is well-suited for intermittent patterns typical of Google Trends queries, improving forecast reliability by accounting for zero-demand intervals. We then used agglomerative hierarchical clustering on principal components (HCPC) derived from Croston-based forecasts to reveal regional structures. HCPC combines dimensionality reduction with clustering, enabling interpretable visualisations of engagement patterns through dendrograms and factor maps (Husson et al., 2010; Lê et al., 2008).

7. Results and Discussion

This section utilises visual analytics to uncover important sub-regional and region-specific insights and yearly trends from the GTE data, using RSI to visualise the dynamic relationship between SDG and AI interests. The graphical analyses highlight spikes in RSI around major SDG events to indicate heightened attention to SDG and statistically detectable AI breakthrough period to indicate rising AI-SDG focus and vice versa. We acknowledge that factors such as youth population, internet and digital infrastructure, economic development (Banga & te Velde, 2018), sector-specific dynamics (HamidiMotlagh et al., 2020), urbanisation (Jedwab et al., 2017), or the technology adoption rate could impact regional differences in how AI solves SDGs, but they are out of the scope of analysis in this study.

7.1. Google Trends Scenarios of AI for Solving SDG in the Central African Region

Over the study period, the Central African region witnessed prolonged periods with little to no SDG search activities since many periods on the time series trends in Figure 1 depict search activities only intermittently. There is evidence of mainly SDG search dominance in all the sub-regions over AI search intensity since Croston’s forecasts for relative salience index (RSI) SDG to AI mostly equal one, highlighting that detectable (spiky) SDG search activity is much higher than AI search activity across all the sub-regions within the periods except in the Republic of Cameroon with some AI dominance from late 2021 through August 2024.

Specifically, the Republic of Cameroon exhibited much higher SDG prominence than AI in November 2015, March 2016, November 2016, August 2017, February and August 2020, and November and December 2023. The region recorded a balance of both SDG and AI search activities in November 2022, but AI search prominence dominated from December 2022 through August 2024 (see Figure A.1 in appendix A).

The remaining Central African sub-regions exhibited pure SDG search dominance over AI search activities, except for the Republic of Chad, which exhibited complete AI search dominance relative to SDG search salience in June 2016 (see Figure A.1 in appendix A). The Republic of Burundi exhibited a dominant SDG search salience over AI only in March 2018 and December 2019, with no AI search dominance over time. The Republic of Gabon had higher relative SDG search salience than AI in only two months over the time periods pre- and post-COVID-19 pandemic in October 2016 and July 2024, with no AI search activity dominance over time. The only detectable (dominant) SDG search activity over AI in the Central African Republic occurred in September 2023, exhibiting a lack of AI search salience beyond.

The Republic of Equatorial Guinea also recorded complete SDG search dominance over AI only recently, in February 2024. The region recorded no AI search activity dominance over the five-year search period. The Republic of Congo (Kinshasa) experienced mainly AI search dominance in June 2021 and July 2024, with SDG search salience only in June 2019. With no AI search dominance over time, the Democratic Republic of Sao Tome and Principe recorded SDG search dominance over AI only in April 2023. Based on time series trend summaries and the empirical evidence from the Central African search records, it is plausible to infer that interest in leveraging AI for solving SDGs is currently low in the Central African sub-region, apart from the Republic of Cameroon. Although both SDG and AI search interests are currently low in the region, there is evidence of higher SDG search activities than AI in the region.

7.2. Google Trends Scenarios of AI for Solving SDG in the Eastern African Region

The trend data of SDG and AI GTE search activities in the Eastern African region is less sparse, with most periods having non-zero values. This dense nature of the data makes the time trend plot of Croston’s forecasts of the RSI cluttered, making it difficult to interpret the relative search salience of AI and SDG across the Eastern African sub-regions. To circumvent this challenge and make the plot less blurry, we applied a curative filter to remove data rows (sub-regions) with search salience estimates outside the 0.3 ≤ RSI ≤ 0.7 range. This filter screens out sub-regions by considering RSI values below 0.3 as low and values from 0.7 upward as high. This makes the study focus on only the seven countries listed in Figure 2 and Figure A.2 in Appendix A, which fall within the most relevant search salience range out of the thirteen Eastern sub-regions.

Of the seven sub-regions within the cut-off range in Figure 2, only the Federal Republic of Somalia lacks an SDG search dominance over AI over time. It, however, recorded AI search dominance over SDG only in early and mid-2022. The Republic of Kenya sub-region exhibited a decreasing SDG search dominance over AI search salience from late 2015 through December-January 2018, after which AI search salience started to gain dominance over SDG search salience through 2024 (see Figure 2 and Figure A.2 in Appendix A). Except in December 2018, when AI dominated SDG, SDG searches were more popular than AI searches in the United Republic of Tanzania and the Republic of Uganda from September 2015 to January 200-2021. Beyond December 2023, AI dominated SDG search salience through August 2024 in both regions. There is evidence of consistent rising AI search salience dominance over SDG search salience in the Federal Democratic Republic of Ethiopia from October 2015 through late 2019. This inference is further evident in the trend patterns of AI to SDG RSI on Figure A.2 in Appendix A.

Furthermore, it is evident in Figure A.2 in Appendix A that the Republic of Mauritius consistently recorded AI dominance over SDG from early 2018 through late 2022. Except in January and March 2022, when SDG search interest was more popular than AI in the Republic of Rwanda, AI search salience dominated SDG salience from mid-2019 through 2024. Overall, SDG- and AI-related search salience is much higher in Eastern Africa than in Central Africa. This inference implies a higher probability of solving SDGs with AI in the Eastern African region than in the Central African region since there were prolonged search activities with detectable balance between AI and SG in the former than in the latter. There is enough data evidence to infer that digital literacy level, access to technological infrastructure, and readiness to leverage innovation for solving SDGs for social good is higher in the Eastern African region.

7.3. Google Trends Scenarios of AI for Solving SDG in the Southern African Region

Croston’s forecast trends for the Southern African region are also less dense in nature, like those of the Eastern African region. We utilised the filter range used on the Eastern region data to the Southern region data to make Figures 3 and A.4 in Appendix A less cluttered, aiding understanding of the trends. The Southern African sub-regions, however, recorded more AI search salience dominance over SDG than in the Eastern sub-regions, where there was an approximate balance.

The Republic of South Africa recorded a consistent AI search interest dominance over SDG search dominance from September 2015 through late 2022 than all other Southern African sub-regions over time. Next to the South African sub-region is the Republic of Zimbabwe regarding AI search dominance over SDG search salience, especially from early 2019 during the COVID-19 pandemic through early 2023. From early 2018 through January 2023, SDG and AI exchanged relative search dominance, after which AI dominance consistently occurred through August 2024. The Republic of Namibia also experienced interchanging SDG and AI search interest dominance from early 2022 through early 2024. The Republic of Botswana only exhibited search activities recently, with SDG dominance in the first few months from late 2021 and mainly AI dominance subsequently through early 2024. The Republic of Malawi mimics the Republic of Botswana, which has had a few months of SDG dominance and balance with AI in 2016 and 2020. The Republic of Mozambique also recently exhibited slightly higher AI search interest from early 2022 through August 2024. Consequently, the Southern African region experienced a much higher SDG and AI search activities over the five years, with significantly higher AI search interest dominance over the SDG salience.

Specifically, there is evidence of a rising trend from COVID-19 in 2019 through 2024 (Ciotti et al., 2020). During COVID-19, from 2019 to 2020, the region exhibited moderately detectable (spiking) AI- and SDG-related search salience trends. Concordance between AI and SDG salience in the region is higher than in other regions. This highlights significant data evidence that current technological innovation regulatory policies and strategies in the region, especially in the South African sub-region, have significantly enhanced AI for solving the SDGs.

7.4. Google Trends Scenarios of AI for Solving SDG in the Northern African Region

The Northern African region experienced more AI search interest dominance than SDG over the five-year study period, as evident from Croston’s search volume forecasts depicted in Figures 4 and A.3 in Appendix A.

Most sub-regions exhibited higher AI search dominance over the study period, except the Western Saharan Republic and Libya, which recorded detectably higher (spiky) SDG search salience than AI in the region in late 2017 and early 2021. It is further evident from the trend plot of AI relative to SDG in Figure A.3 in Appendix A that AI search salience is significantly higher (RSI = 1.0) than SDG salience (RSI ≤ 0.7) in all sub-regions except Libya and Western Sahara over time.

7.5. Google Trends Scenarios of AI for Solving SDG in the Western African Region

The Western African region data contained more SDG and AI search activities than in the Central and Northern African regions, with evident consistent switching SDG and AI search interest dominance trends in the Federal Republic of Nigeria and the Republic of Ghana in Figures 5 and A.5. Specifically, the Republic of Ghana started with a higher SDG search interest dominance over AI ahead of the Federal Republic of Nigeria from September 2015 through August 20204, while the latter sub-region got ahead of the former with consistently higher AI search salience over SDG through August 2024.

In addition to the Nigerian and Ghana sub-regions, only six other sub-regions, including the Republic of Sierra Leone, the Republic of Benin, the Republic of Senegal, the Republic of Guinea, Burkina Faso, and the Republic of Liberia, exhibited detectable (spiky) SDG search interest dominance over AI over time. Of these six additional sub-regions with SDG dominance, only the Republic of Sierra Leone exhibited consistent dominance over time, with an approximate SDG-AI salience balance in the later months of 2023. The remaining six Western African sub-regions, including the Republic of Cabo Verde (Cape Verde), the Republic of the Gambia, the Republic of Guinea-Bissau, the Republic of Mali, the Republic of Niger, and the Togolese Republic, had approximately zero SDG and AI search activities over time.

The empirical evidence in this study confirms the plausibility of growing interest in harnessing AI for solving SDGs for social good in the Federal Republic of Nigeria and the Republic of Ghana. Other sub-regions currently belong to two distinct groups: those with no preparedness and those intermittently SDG-focused only. This situation presents an overarching need for intentional AI adoption strategies and financial inclusion-backed digital literacy to drive and enhance the implementation of the SDGs in the region.

8. Regional Patterns in AI and SDG Search Volumes

As a follow-up on the analytics of the AI for solving SDG trends in the previous sections, we performed a hierarchical cluster analysis on the regional search trends data to discover the hidden search salience patterns among the sub-regions. The analytics present cluster analysis dendrograms for each region to reduce the complexity of the individual trends, simplifying decision-making for policymakers and stakeholders.

8.1. Sub-region Clusters with Central Africa

The hierarchical clustering dendrogram in Figure 6 reflects four significant clusters of the Central Africa sub-regions regarding AI- and SDG-related search interest volumes over the five-year period under investigation. The Gabonese Republic and the Republics of Cameroon and Chad occupied distinct clusters from the rest of the sub-regions.

The Republic of Cameroon is distinctively popular in terms of AI and SDG dominance over other sub-regions, while the Republic of Chad is distinctively popular in terms of AI search interests. The Gabonese Republic in the third cluster is Popular for SDG search interests, with other sub-regions achieving low AI and SDG search interests in the fourth cluster.

8.2. Sub-region Clusters with Eastern Africa

The clustering evidence in Figure 7 reveals three distinct groupings of the Eastern sub-regions regarding possible interests in adopting AI for solving SDG. The Republic of Kenya and the United Republic of Tanzania in cluster two (from the right) comprise the most popular Eastern sub-regions regarding solving SDG with AI, as evidenced in the RSI analytics in Figure 2 and A.2. The Federal Democratic Republic of Ethiopia and the Republics of Mauritius, Rwanda, and Uganda belong to the second most popular sub-regions on solving SDG with AI in the Eastern region belong to the first cluster.

The remaining sub-regions in the third cluster exhibited non-detectable search interests for solving SDG with AI over the nine years under consideration.

8.3. Sub-region Clusters with Northern Africa

In furtherance of the inference from the time series RSI plot in Figures 3 and A.3, it is evident that the Arab Republic of Egypt and the Western Sahara Republic belong to the first cluster in Figure 8 as the most popular Northern sub-regions with significant interest in leveraging AI for solving SDG. The People’s Democratic Republic of Algeria, the Republic of Tunisia, and the Kingdom of Morocco belong to the second cluster, while Libya and the Islamic Republics of Mauritania are in the third cluster. The second cluster sub-regions exhibited higher SDG dominance than AI, but there was more AI dominance in the third cluster sub-regions over the study period.

8.4. Sub-region Clusters with Southern Africa

The salience of AI for addressing SDGs in the Southern African sub-regions is categorised into the five distinct clusters in the dendrogram in Figure 9. The Republics of South Africa and Zimbabwe in cluster five (from the right) are the most popular sub-regions in Southern Africa in terms of consistency in AI for solving SDG dominance over time and their RSI concordance. The Republic of Malawi, in the fourth cluster, ranks second in terms of potential readiness for solving SDG with AI.

In the second cluster, the Republics of Zambia and Namibia rank third as the go-to countries for solving SDGs with AI in the Southern African Region. According to clustering and RSI analytics insights, the Republic of Botswana ranks fourth in the first cluster. The Kingdoms of Eswatini and Lesotho, and the Republics of Angola and Mozambique in the third cluster rank fifth in the region as the group of countries with statistically non-detectable AI for solving SDG readiness due to mostly zero AI and SDG search activity records over the time.

8.5. Sub-region Clusters with Western Africa

The empirical evidence from the AI and SDG search trends RSI in Figures 5, A.5, and 10 reflected that the Federal Republic of Nigeria and the Republic of Ghana in the first cluster are most comparably popular in solving SDG with AI in the Western African region. The colour difference between the former and the latter sub-regions in cluster one reflects that they are not a perfect match, with Nigeria having higher search salience. The Republics of Sierra Leone and Senegal, in the fourth and the fifth clusters in Figure 10, rank second and third in terms of AI for solving SDG readiness due to their detectable search volumes and RSI trends. Regarding AI search salience with little SDG search activities, the Republic of Liberia and the Togolese Republic rank fourth in the sixth cluster. For SDG search salience, the Republic of Mali in the second cluster ranks fifth. The remaining sub-regions in the fourth cluster rank sixth due to few detectable search activities over the five years under review.

9. Challenges and Limitations

While Google Trends provides valuable insights, there are several limitations to using this tool for analysing AI and SDG trends in Africa. One major limitation is the unequal distribution of internet access across African countries, particularly between urban and rural areas. This can introduce bias into the results, as countries with higher internet penetration may show inflated search interest compared to those with limited access. Additionally, Google Trends does not provide detailed demographic data, which could limit the depth of analysis of user behaviour and interests. Finally, search interest data only reflects online behaviour, which may not always correlate with real-world adoption or progress toward achieving the SDGs. Despite these limitations, Google Trends remains useful for capturing public interest and engagement with AI and sustainable development topics.

10. Conclusion

This study has highlighted valuable insights into AI adoption’s relative salience indices for solving the SDGs, the current adoption readiness rankings, and the differences in regional trends across Africa. The regional analytics of trend scenarios revealed that the relative search salience of AI and SDG is on the rise in Africa, with AI search interests mostly dominating SDG searches. All five African regions have witnessed dominating AI search interest salience across their sub-regions over the past nine-year period from September 2015 through August 2024, except in the Central African region, where SDG search salience mostly dominated over AI salience. It is plausible to infer from the empirical evidence in this study that interest in leveraging AI for solving SDGs is currently low in the Central African sub-region, except in the Republic of Cameroon.

The Republic of Kenya and the United Republic of Tanzania comprise the most popular sub-regions for solving SDGs with AI in the Eastern African region. The Arab Republic of Egypt and the Western Sahara Republic are the most popular Northern sub-regions interested in leveraging AI to solve SDGs. The Republics of South Africa and Zimbabwe are the most popular sub-regions in Southern Africa and across the African continent regarding consistency in AI for solving SDG dominance over time and their RSI concordance. The Federal Republic of Nigeria and the Republic of Ghana are most comparably popular in terms of solving SDG with AI in the Western African region.

Over the past nine-year period, the Central African region had the highest number of countries with statistically non-detectable AI for solving SDG readiness due to mostly zero AI and SDG search activity records. The countries include the Republic of Burundi, the Republic of the Congo (Kinshasa), the Democratic Republic of Congo (Brazzaville), the Democratic Republic of São Tomé and Príncipe, the Republic of Equatorial Guinea, and the Central African Republic. In the Eastern African Region, the Republic of South Sudan, the Republic of Sudan, the Republic of Madagascar, the State of Eritrea, the Republic of Seychelles, the Republic of Djibouti, and the Federal Republic of Somalia exhibited non-detectable search interests for solving SDG with AI over the nine years under consideration. The Republic of Tunisia, the People’s Democratic Republic of Algeria, and the Kingdom of Morocco exhibited higher SDG dominance than AI search interests in the Northern Africa region over time. The Kingdoms of Eswatini and Lesotho and the Republics of Angola and Mozambique in Southern Africa exhibited statistically non-detectable AI for solving SDG readiness due to mostly zero AI and SDG search activity records over time. The empirical insights from this study have detected these countries to be backwards regarding readiness to leverage AI for solving the SDGs, highlighting the overarching need for intentional policy reviews and strategic planning to improve digital literacy and technological infrastructure development by the government and relevant stakeholders to raise AI and SDG awareness.

Based on the empirical insights, it is plausible that the Southern, Western, Eastern, Northern, and Central African regions ranked first through fifth on their current adoption readiness rankings for solving SDGs with AI. The current differences in regional trends across the African regions are multi-dimensional. They can be better simplified by supplementing GTE search data with additional indicators such as GDP per capita, digital literacy rates, internet penetration, and national policy frameworks on AI and sustainable development. These indicators would provide a richer context for interpreting readiness and engagement patterns and help explain underlying socioeconomic or structural drivers that influence public interest in AI and SDGs.

References

Ade-Ibijola, A., & Okonkwo, C. (2023). Artificial intelligence in Africa: Emerging challenges. In Responsible AI in Africa: Challenges and Opportunities (pp. 101-117). Springer International Publishing Cham. https://doi.org/10.1007/978-3-031-08215-3_5

African Union. (2020). The Digital Transformation Strategy For Africa (2020-2030). A. Union. Retrieved from: https://tinyurl.com/yc8b38jb

African Union. (2024a). Continental Artificial Intelligence Strategy: Harnessing AI for Africa’s Development and Prosperity. A. Union. Retrieved from: https://tinyurl.com/s3tmk9w4

African Union. (2024b). Member States. Retrieved September 05, 2024 from https://au.int/en/member_states/countryprofiles2

Ajaj, R., Buheji, M., & Hassoun, A. (2024). Optimizing the readiness for Industry 4.0 in fulfilling the Sustainable Development Goal 1: Focus on poverty elimination in Africa. Frontiers in Sustainable Food Systems, 8, 1393935. https://doi.org/10.3389/fsufs.2024.1393935

Arfanuzzaman, M. D. (2021). Harnessing artificial intelligence and big data for SDGs and prosperous urban future in South Asia. Environmental and sustainability indicators, 11, 100127. https://doi.org/10.1016/j.indic.2021.100127

Arthur, K. K., Asongu, S. A., Darko, P., Ansah, M. O., Adom, S., & Hlortu, O. (2024). Financial crimes in Africa and economic growth: Implications for achieving sustainable development goals (SDGs). Journal of Economic Surveys. https://doi.org/10.1111/joes.12652

Astobiza, A. M., Toboso, M., Aparicio, M., & López, D. (2021). AI ethics for sustainable development goals. IEEE Technology and Society Magazine, 40(2), 66-71. https://doi.org/10.1109/MTS.2021.3056294

Banga, K., & te Velde, D. W. (2018). Digitalisation and the Future of Manufacturing in Africa. ODI London. https://tinyurl.com/jy7h6v3n

Bjola, C. (2022). AI for development: Implications for theory and practice. Oxford Development Studies, 50(1), 78-90. https://doi.org/10.1080/13600818.2021.1960960

Choi, H., & Varian, H. (2012). Predicting the present with Google Trends. Economic record, 88, 2-9. https://doi.org/10.1111/j.1475-4932.2012.00809.x

Chui, M., Manyika, J., Miremadi, M., Henke, N., Chung, R., Nel, P., & Malhotra, S. (2018). Notes from the AI frontier: Insights from hundreds of use cases. McKinsey Global Institute, 2, 267. https://tinyurl.com/3dcwyaa5

Ciotti, M., Ciccozzi, M., Terrinoni, A., Jiang, W. C., Wang, C. B., & Bernardini, S. (2020). The COVID-19 pandemic. Critical Reviews in Clinical Laboratory Sciences, 57(6), 365–388. https://doi.org/10.1080/10408363.2020.1783198

Croston, J. D. (1972). Forecasting and Stock Control for Intermittent Demands. Journal of the Operational Research Society, 23(3), 289-303. https://doi.org/10.1057/jors.1972.50

Deloitte. (2025). AI trends 2025: Adoption barriers and updated predictions. https://www.deloitte.com/us/en/services/consulting/blogs/ai-adoption-challenges-ai-trends.html

Fazal, A., Ahmed, A., & Abbas, S. (2024). Importance of artificial intelligence in achieving sustainable development goals through financial inclusion. Qualitative Research in Financial Markets. https://doi.org/10.1108/QRFM-04-2023-0098

Franzén, A. (2023). Big data, big problems: Why scientists should refrain from using Google Trends. Acta Sociologica, 66(3), 343-347. https://doi.org/10.1177/00016993221151118

Google Cloud. (2025). AI Business Trends 2025. https://cloud.google.com/resources/ai-trends-report

Google Research Brief. (2024). AI in Action: Accelerating Progress Towards the Sustainable Development Goals. https://static.googleusercontent.com/media/publicpolicy.google/en//resources/research-brief-ai-and-SDG.pdf

Google Trends. (2024). Google Trends Explore Data. Retrieved August 05, 2024 from https://www.google.com/trends

Goralski, M. A., & Tan, T. K. (2020). Artificial intelligence and sustainable development. The International Journal of Management Education, 18(1), 100330. https://doi.org/10.1016/j.ijme.2019.100330

Grekov, A. N., Vyshkvarkova, E. V., & Mavrin, A. S. (2024). Forecasting and Anomaly Detection in BEWS: Comparative Study of Theta, Croston, and Prophet Algorithms. Forecasting, 6(2), 343. https://doi.org/10.3390/forecast6020019

HamidiMotlagh, R., Babaee, A., Maleki, A., & Taghi Isaai, M. (2020). Innovation policy, scientific research and economic performance: The case of Iran. Development Policy Review, 38(3), 387-407. https://doi.org/10.1111/dpr.12423

Hoang, T. G., Nguyen, G. N. T., & Le, D. A. (2022). Developments in financial technologies for achieving the Sustainable Development Goals (SDGs): FinTech and SDGs. In Disruptive technologies and eco-innovation for sustainable development (pp. 1-19). IGI Global. https://doi.org/10.4018/978-1-7998-8900-7.ch001

Holzinger, A., Weippl, E., Tjoa, A. M., & Kieseberg, P. (2021). Digital transformation for sustainable development goals (sdgs)-a security, safety and privacy perspective on ai International cross-domain conference for machine learning and knowledge extraction. https://doi.org/10.1007/978-3-030-84060-0_1

Hoosain, M. S., Paul, B. S., & Ramakrishna, S. (2020). The impact of 4ir digital technologies and circular thinking on the united nations sustainable development goals [Article]. Sustainability (Switzerland), 12(23), 1-16, Article 10143. https://doi.org/10.3390/su122310143

Husson, F., Lê, S., & Pagès, J. (2010). Exploratory multivariate analysis by example using R [Book]. https://doi.org/10.1201/b10345

Jedwab, R., Christiaensen, L., & Gindelsky, M. (2017). Demography, urbanization and development: Rural push, urban pull and… urban push? Journal of Urban Economics, 98, 6-16.

Jun, S.-P., Yoo, H. S., & Choi, S. (2018). Ten years of research change using Google Trends: From the perspective of big data utilizations and applications. Technological forecasting and social change, 130, 69-87. https://doi.org/10.1016/j.techfore.2017.11.009

Keesara, S., Jonas, A., & Schulman, K. (2020). Covid-19 and health care’s digital revolution. New England Journal of Medicine, 382(23), e82. https://doi.org/10.1056/NEJMp2005835

Kourentzes, N. (2014). On intermittent demand model optimisation and selection. International Journal of Production Economics, 156, 180-190. https://doi.org/10.1016/j.ijpe.2014.06.007

Lê, S., Josse, J., & Husson, F. (2008). FactoMineR: An R package for multivariate analysis [Article]. Journal of Statistical Software, 25(1), 1-18. https://doi.org/10.18637/jss.v025.i01

Mavragani, A., Ochoa, G., & Tsagarakis, K. P. (2018). Assessing the methods, tools, and statistical approaches in Google Trends research: systematic review. Journal of Medical Internet Research, 20(11), e270.

Meitei, A. J., Rai, P., & Rajkishan, S. S. (2023). Application of AI/ML techniques in achieving SDGs: a bibliometric study. Environment, Development and Sustainability, 1-37. https://doi.org/10.1007/s10668-023-03935-1

Mhlanga, D. (2020). Artificial Intelligence (AI) and poverty reduction in the Fourth Industrial Revolution (4IR). https://doi.org/10.20944/preprints202009.0362.v1

Mhlanga, D. (2021). Artificial intelligence in the industry 4.0, and its impact on poverty, innovation, infrastructure development, and the sustainable development goals: Lessons from emerging economies? Sustainability, 13(11), 5788. https://doi.org/10.3390/su13115788

Mukonza, S. S., & Chiang, J. L. (2023). Meta-Analysis of Satellite Observations for United Nations Sustainable Development Goals: Exploring the Potential of Machine Learning for Water Quality Monitoring [Review]. Environments - MDPI, 10(10), Article 170. https://doi.org/10.3390/environments10100170

Pedro, F., Subosa, M., Rivas, A., & Valverde, P. (2019). Artificial intelligence in education: Challenges and opportunities for sustainable development.

Prestwich, S. D., Tarim, S. A., & Rossi, R. (2021). Intermittency and obsolescence: A Croston method with linear decay. International Journal of Forecasting, 37(2), 708-715. https://doi.org/10.1016/j.ijforecast.2020.08.010

Raman, R., Sreenivasan, A., Ma, S., Patwardhan, A., & Nedungadi, P. (2023). Green Supply Chain Management Research Trends and Linkages to UN Sustainable Development Goals. Sustainability, 15(22), 15848. https://doi.org/10.3390/su152215848

Ramezani, M., Takian, A., Bakhtiari, A., Rabiee, H. R., & Sazgarnejad, S. (2023). Bibliometric Analysis of Artificial Intelligence Revolutions in Health-related Sustainable Development Goals [Article]. Health Technology Assessment in Action, 7(4). https://doi.org/10.18502/htaa.v7i4.14654

Roberts, M., Driggs, D., Thorpe, M., Gilbey, J., Yeung, M., Ursprung, S., Aviles-Rivero, A. I., Etmann, C., McCague, C., & Beer, L. (2021). Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scans. Nature Machine Intelligence, 3(3), 199-217. https://doi.org/10.1038/s42256-021-00307-0

Roy, A., Basu, A., Su, Y., Li, Y., & Dong, X. (2022). Understanding recent trends in global sustainable development goal 6 research: scientometric, text mining and an improved framework for future research. Sustainability, 14(4), 2208. https://doi.org/10.3390/su14042208

Salvia, A. L., Leal Filho, W., Brandli, L. L., & Griebeler, J. S. (2019). Assessing research trends related to Sustainable Development Goals: Local and global issues. Journal of Cleaner Production, 208, 841-849. https://doi.org/10.1016/j.jclepro.2018.09.242

Sampene, A. K., Agyeman, F. O., Robert, B., & Wiredu, J. (2022). Artificial intelligence as a path way to Africa’s transformations. Artificial Intelligence, 9(1).

Schultz, C. R. (1987). Forecasting and Inventory Control for Sporadic Demand Under Periodic Review. Journal of the Operational Research Society, 38(5), 453-458. https://doi.org/10.1057/jors.1987.74

Shafi, I., Sohail, A., Ahmad, J., Espinosa, J. C. M., López, L. A. D., Thompson, E. B., & Ashraf, I. (2023). Spare parts forecasting and lumpiness classification using neural network model and its impact on aviation safety. Applied Sciences, 13(9), 5475. https://doi.org/10.3390/app13095475

Sinde, R., Diwani, S., Leo, J., Kondo, T., Elisa, N., & Matogoro, J. (2023). AI for Anglophone Africa: Unlocking its adoption for responsible solutions in academia-private sector [Article]. Frontiers in Artificial Intelligence, 6, Article 1133677. https://doi.org/10.3389/frai.2023.1133677

Singh, A., Kanaujia, A., Singh, V. K., & Vinuesa, R. (2024). Artificial intelligence for Sustainable Development Goals: Bibliometric patterns and concept evolution trajectories. Sustainable Development, 32(1), 724-754. https://doi.org/10.1002/sd.2706

Singh, S., & Verma, S. B. (2024). Resolving Covid-19 with Blockchain and AI: A Systematic Review. ADCAIJ: Advances in Distributed Computing and Artificial Intelligence Journal, 13(1), e31454. https://doi.org/10.14201/adcaij.31454

Snyder, R. (2002). Forecasting sales of slow and fast moving inventories. European Journal of Operational Research, 140(3), 684-699. https://doi.org/10.1016/S0377-2217(01)00231-4

Steingard, D., Balduccini, M., & Sinha, A. (2023). Applying AI for social good: Aligning academic journal ratings with the United Nations Sustainable Development Goals (SDGs). AI & SOCIETY, 38(2), 613-629. https://doi.org/10.1007/s00146-022-01459-2

Svetunkov, I., & Boylan, J. E. (2023). iETS: State space model for intermittent demand forecasting. International Journal of Production Economics, 265, 109013. https://doi.org/10.1016/j.ijpe.2023.109013

Teunter, R. H., Syntetos, A. A., & Zied Babai, M. (2011). Intermittent demand: Linking forecasting to inventory obsolescence. European Journal of Operational Research, 214(3), 606-615. https://doi.org/10.1016/j.ejor.2011.05.018

Truby, J. (2020). Governing artificial intelligence to benefit the UN sustainable development goals. Sustainable Development, 28(4), 946-959. https://doi.org/10.1002/sd.2048

UNESCO. (2022). AI policies in South Africa. UNESCO. Retrieved 1/11/2024 from: https://oecd.ai/en/dashboards/countries/SouthAfrica

UNESCO. (2024). Shaping Kenya’s AI Future: UNESCO Contributes to National AI Strategy Formulation. UNESCO. Retrieved 11/11/2024 from: https://www.unesco.org/en/articles/shaping-kenyas-ai-future-unesco-contributes-national-ai-strategy-formulation

United Nations. (2015). Transforming our world: The 2030 agenda for sustainable development. United Nations. Retrieved 11/11/2024 from https://sustainabledevelopment.un.org/post2015/transformingourworld

Vinuesa, R., Azizpour, H., Leite, I., Balaam, M., Dignum, V., Domisch, S., Felländer, A., Langhans, S. D., Tegmark, M., & Fuso Nerini, F. (2020). The role of artificial intelligence in achieving the Sustainable Development Goals [Review]. Nature Communications, 11(1), Article 233. https://doi.org/10.1038/s41467-019-14108-y

Wakunuma, K., Ogoh, G., Eke, D. O., & Akintoye, S. (2022). Responsible AI, SDGS, and AI governance in Africa 2022. IST-Africa Conference (IST-Africa). https://doi.org/10.23919/IST-Africa56635.2022.9845598

Appendix A