Sustainable development in the age of AI

As AI revolutionizes research, healthcare and industry, these technologies also have the potential to advance and monitor the UN Sustainable Development Goals (SDGs) abre em uma nova guia/janela. To accelerate SDG outcomes, however, regulation, governance and effective prioritization of AI are fundamental.

Elsevier’s recent Insights 2024 survey suggests there is currently insufficient regulation or governance for AI. This sentiment is echoed by a significant number of researchers across the globe who also recognize that AI can accelerate knowledge discovery.

To delve more deeply into this topic, we interviewed three scholars whose research is showcased in Elsevier’s special collection In the Age of AI: Information to Support UN SDGs abre em uma nova guia/janela.

They talked about how AI technologies are being used to advance and monitor the SDGs, how AI can enhance decision-making, the potential for scaling up AI efforts, and the predicted impact of AI on SDG outcomes.

Here are their key recommendations:

• Researchers and policymakers should strive to develop and implement AI technologies that take all 17 SDGs into account.

• Building IT infrastructure and technological readiness is crucial for success, especially for low- and lower-middle-income countries.

• Stronger local and international AI governance is needed, and this governance should be transparent, inclusive, people-centric and grounded in ethics.

• AI developers must be transparent about their technologies, use quality and up-to-date data to train their models, and incorporate measures to ensure data privacy and security.

Develop AI technologies that take all SDGs into account

Although AI technologies are already being used to tackle all 17 SDGs, AI innovation and policy are currently skewed towards SDGs related to economic growth. By analyzing the current global use of AI in research and field projects, Prof Junaid Qadir abre em uma nova guia/janela and colleagues showed that the development of AI technology and ethics frameworks are mainly focused on improving economic growth.¹ Over 50% of the sources they analyzed were concerned with industry, innovation and infrastructure (SDG9), while less than 5% focused on gender equality, clean water and sanitation, life below water, and life on land (SDGs 5, 6, 14 and 15).

Given the gaps in AI development for SDGs, how much impact is AI predicted to have on SDG outcomes? Though many studies have investigated how AI is currently being used, Dr Sharmin Nahar abre em uma nova guia/janela, Associate Professor of Entrepreneurship at Lancaster University abre em uma nova guia/janela, is the first to forecast the impact of AI technology on SDG outcomes. Her simulations predict that AI technologies will significantly reduce poverty, improve health and wellbeing, and improve gender equality (SDGs 1, 3 and 5) but will not meaningfully improve responsible consumption and production, life below water, or life on land (SDGs 12, 14 and 15) for most countries, based on current efforts.²

Improve AI readiness and investing in IT infrastructure

All three experts emphasize that improving technological readiness is critical to maximizing the benefits of AI technologies. To achieve this, countries need to be open to technological change, have a foundation of IT infrastructure and internet access, and be able to collect and store quality data to measure progress.

In his research, Md Arfanuzzaman abre em uma nova guia/janela focuses on South Asian countries. He notes that while the region is poised to harness AI technologies for SDG monitoring and implementation due to its rapid economic growth and young, tech-savvy population, it faces unique social and environmental challenges.³ He says more investment in IT infrastructure and data collection could accelerate progress in achieving the region’s goals and emphasizes the importance of closing the technological divide between urban and rural areas.

On a global scale, Dr Nahar emphasizes the importance of international investment and collaboration in AI-driven innovation, highlighting the need for high-quality and accessible data. “To understand the impact of AI on SDGs and to enhance evidence-based decision-making, we need to collect high-quality empirical data from individual countries,” she says.

Strengthen global cooperation and governance

Because SDGs represent global goals and AI technologies have a global impact, international cooperation, governance and ethical oversight are critical. The researchers emphasize that this governance needs to occur across scales — from local to global.

“We need to have agreement among all humanity on the basics but allow room for local aspects, as rules that complement local values, culture and religion are more likely to be accepted,” Prof Qadir explains.

In addition to international cooperation, he says, collaborations among the private and public sectors and partnerships between academia, industry and government will be crucial for AI technologies to effectively support sustainable development goals.

Recognize that AI is a mirror, not a panacea

“As we look at AI for SDGs, we must examine how sustainable our own behavior and value systems are,” says Prof Qadir. “If our notions of progress are not sustainable, no AI can help us because AI just mirrors whatever we show value for.”

At the end of the day, AI’s potential to support sustainable development comes down to how humanity wields it.