[Paper Review] AI in Action: Accelerating Progress Towards the Sustainable Development Goals

Brigitte Hoyer Gosselink, Kate Brandt+9 more|arXiv (Cornell University)|Jul 2, 2024

Economic and Technological Innovation7 citations

TL;DR

The paper examines how AI can accelerate progress on the UN SDGs, highlighting capabilities and use cases across all 17 goals with a focus on health, education, and climate, and provides guidance for responsible AI deployment.

ABSTRACT

Advances in Artificial Intelligence (AI) are helping tackle a growing number of societal challenges, demonstrating technology's increasing capability to address complex issues, including those outlined in the United Nations (UN) Sustainable Development Goals (SDGs). Despite global efforts, 80 percent of SDG targets have deviated, stalled, or regressed, and only 15 percent are on track as of 2023, illustrating the urgency of accelerating efforts to meet the goals by 2030. We draw on Google's internal and collaborative research, technical work, and social impact initiatives to show AI's potential to accelerate action on the SDGs and make substantive progress to help address humanity's most pressing challenges. The paper highlights AI capabilities (including computer vision, generative AI, natural language processing, and multimodal AI) and showcases how AI is altering how we approach problem-solving across all 17 SDGs through use cases, with a spotlight on AI-powered innovation in health, education, and climate. We then offer insights on AI development and deployment to drive bold and responsible innovation, enhance impact, close the accessibility gap, and ensure that everyone, everywhere, can benefit from AI.

Motivation & Objective

• Motivate the urgency to accelerate SDG progress given that 80% of targets have deviated, stalled, or regressed and only 15% are on track as of 2023.
• Demonstrate AI's potential to address complex SDG challenges using Google's internal and collaborative research.
• Showcase AI capabilities (vision, generative AI, NLP, multimodal) and problem-solving approaches across all 17 SDGs.
• Highlight AI-driven use cases in health, education, and climate to illustrate actionable impact.
• Provide insights on development and deployment practices to drive bold, responsible innovation and broader accessibility.

Proposed method

• Leverages Google's internal research, collaborative studies, technical work, and social impact initiatives.
• Describes AI capabilities such as computer vision, generative AI, natural language processing, and multimodal AI.
• Connects AI capabilities to SDG problem-solving across all 17 goals through practical use cases.
• Spotlights AI-powered innovations in health, education, and climate to illustrate impact.
• Proposes guidelines for responsible AI development and deployment to close accessibility gaps.
• Uses evidence from internal and collaborative projects to argue for scalable, inclusive AI solutions.

Experimental results

Research questions

• RQ1 How can AI technologies accelerate progress toward achieving the UN Sustainable Development Goals?
• RQ2 Which AI capabilities and applications produce the most impact across different SDGs?
• RQ3 What deployment and governance practices are needed to ensure responsible, accessible, and broadly beneficial AI for SDGs?
• RQ4 What lessons from health, education, and climate use cases inform scalable AI-driven action toward the SDGs.

Key findings

• 80 percent of SDG targets have deviated, stalled, or regressed, underscoring the urgency to accelerate progress.
• Only 15 percent of SDG targets are on track as of 2023, highlighting a critical gap to close by 2030.
• AI capabilities such as computer vision, generative AI, NLP, and multimodal AI enable new approaches to SDG problem-solving.
• Use cases in health, education, and climate illustrate how AI can catalyze action across SDGs.
• The paper discusses development and deployment insights to drive bold, responsible innovation and close accessibility gaps.
• The findings emphasize ensuring that AI benefits are widespread, ethical, and inclusive.