The United Nations’ Sustainable Development Goals (SDGs) aim to eradicate poverty, protect the environment, combat climate change, and bolster global peace and prosperity by the year 2030. Despite extensive research, additional work is required to accurately forecast SDG scores, which measure progress towards these objectives. By employing ARIMAX and Linear Regression machine learning models which are smoothed using the Holt-Winters’ multiplicative method, the current study predicts SDG scores for various global regions by 2030 based on past data. These forecasting models include predictors, influenced by future use of artificial intelligence (AI), and derived from particular sustainable goals related to health, education, clean energy, and climate action.
AI can provide significant boosts to progress in these key SDGs by reducing energy consumption, improving environmental monitoring, and enhancing health and education sectors. However, potential risks and challenges also arise, such as privacy violations, increased inequality, and technological unemployment. To control these potential adverse effects, careful regulation and international guidelines must be put in place.
Findings from the models illustrate that OECD countries, followed by Eastern Europe, Central Asia, Latin America, and the Caribbean, are anticipated to achieve the highest SDG scores by 2030. Meanwhile, East and South Asia, the Middle East, North Africa, and Sub-Saharan Africa are predicted to improve but remain lower in performance. It is therefore crucial for policymakers to leverage the benefits of AI to uplift regions that are lagging in SDG achievement, while also taking into account socio-economic, cultural, and political factors influencing such progress.
Despite these insights, certain limitations persist. These include the inherent uncertainty in model-based predictions and the continually growing and unpredictable influence of AI. Future research should therefore aim to refine these forecasting models by incorporating a broader range of economic, social, and environmental predictors. Additionally, assessing the influence of policy changes on SDG outcomes should be a focus in ongoing research.
In conclusion, the study demonstrates the potential of machine learning models in predicting SDG scores for global regions by 2030, showcasing a trend of overall improvement albeit with variation across regions. By emphasising the role of AI in accelerating the achievement of SDGs, it underscores the need for careful management of technological advancements in tandem with targeted policy efforts to bolster sustainable development.
