Following the public launch of ChatGPT and the plethora of AI-powered tools emerging on what feels like a weekly basis, the opportunities to harness this technology appear limitless. Some fear that SkyNet’s rule of an enslaved humanity is imminent, others are excited about AI as an economic growth catalyst. Governments are scrambling to keep up with the mind-blowing pace of tech development and establish some form of regulatory frameworks (e.g. European Union’s 2024 Artificial Intelligence Act).
The specific question I pose here is: how can we use AI to accelerate progress towards the UN Sustainable Development Goals (SDGs)?
The following is an updated extract of a 2021 essay I wrote as part of my M.Sc. studies at Kiel University. This may sound too ‘old’ in this fast-moving space but as it examines underlying structures instead of specific tech solutions I regard it to be of ongoing relevance: It provides a brief introduction to the six key dimensions of sustainability that play a role in technological developments and how AI could support the delivery of specific UN SDGs, namely SDG 3 Good Health and Wellbeing, SDG 5 Gender Equality, SDG 10 Reduce Inequality and SDG 16 Peace, Justice and strong institutions.
AI’s Contribution to Sustainability
One of the most widely used definitions surrounding sustainability from the Brundtland Report namely, “Sustainable development is a development that meets the needs of the present without compromising the ability of future generations to meet their own needs.” (Brundtland, 1987). However, sustainability encompasses much more, especially when speaking of technological developments.
According to Becker et al. (2016), there are six key dimensions of sustainability that play a role in technological developments:
- Individual: individual freedom, agency, and dignity
- Social: mutual trust and communication in a social system
- Economic: financial stability and value creation
- Technical: maintaining and improving artificial systems
- Environmental: use and the stewardship of the natural environment
On the individual dimension, AI can benefit individuals and companies by automating tasks that significantly reduce working hours, thus improving workers’ well-being (Khakurel et al., 2018). The rise of AI technology can also lead to distress in individuals for fear of job loss or privacy infringement (Khakurel et al., 2018).
On the social dimension, AI may act as an enabler for all the targets by supporting the provision of food, health, water, and energy services. It can also underpin low-carbon systems, for instance, by supporting the creation of circular economies and smart cities that efficiently use their resources (Vinuesa et al., 2020). Moreover, AI can strengthen communities, as it can support networking, administration, and collaboration (Khakurel et al., 2018).
On the economic dimension, AI technologies can reduce long-term production costs with more reliable systems, improve supply chain efficiency, and flexibility (Vinuesa et al., 2020). However, according to Khakurel et al. (2018), these benefits will only occur in wealthier nations, implementing this technology. Moreover, low and middle-skilled workers will be pressured to acquire new skills that cannot be conducted by AI or be left without employment. For this to occur, Khakurel et al. (2018) argue that questions of trust and liability regarding AI must be transparent; if they aren’t, it can lead to severe conflicts and damage.
On the technical dimension, AI can immensely contribute to fast technological advancements in all sectors. Yet Khakurel et al. (2018) emphasize that the level of sustainability here is highly dependent on the designer and their ethical codes. Thus, they recommend more development and implementation of said values to ensure sustainable development.
On the environmental dimension, AI can advance the understanding of climate change and the modeling of its possible impacts (Vinuesa et al., 2020). Moreover, it can be used to prevent and significantly reduce marine pollution of all kinds. Algorithms can be used for automatic identification of possible oil spills, combating desertification, and restoring degraded land and soil (Khakurel et al., 2018). AI is also increasingly applied to improve biodiversity monitoring and conservation (Vinuesa et al., 2020). However, AI can also negatively impact the environment. It will increase the production and consumption of goods, leading to the perpetuation of planned obsolescence and the further depletion of natural resources (Khakurel et al., 2018).
AI’s Relationship to the Sustainable Development Goals
Artificial Intelligence (AI) is predicted to have an impact on the Sustainable Development Goals (SDG), with recent research suggesting that it will have a positive effect on 134 targets across the different goals and will harm 59 of the targets across the different goals (Vinuesa et al., 2020). We will explore the potential impact of AI on four of these goals: SDG 3 (good health and well-being); SDG 5 (gender equality); SDG 10 (reducing inequality); and SDG 16 (peace, justice, and strong institutions).
SDG 3: Good Health and Wellbeing
SDG 3 concerns good health and wellbeing, including reducing the global incidence of disease, decreasing maternal mortality rates, and increasing life expectancy (UNDP, 2020). Yet there are many challenges to achieving good health for all of humankind, not least the lack of information and knowledge about disease and conditions in certain regions. AI could help ease some of these challenges: for example, data-driven AI technologies and techniques could become one of the primary medicines used in the fight against disease and the prevention of the spread of illness (IISD, 2019). Big data, coupled with AI, could be combined to offer remote care and mobile health options, which could revolutionize medicine and medical care (Benka, 2018). These considerations are especially important in developing countries where medical facilities are scarce, but high percentages of people have access to mobile phones and smartphone technologies, including apps (DAIA, 2019). There is great potential for AI to positively contribute to the fulfillment of the goals and targets in SDG 3.
SDG 5: Gender Equality
SDG 5 concerns gender equality, the aim to end all discrimination against women and girls, to establish fundamental human rights and, from that point, ensure a sustainable, gender-equitable, future (UNDP, 2019). The main targets of SDG 5 include eliminating discrimination and violence against women and girls, eliminating harmful practices against women and girls (including forced marriage and genital mutilation), and ensuring full participation in democratic processes, amongst other targets (UNDP, 2019).
The World Economic Forum (2019) points out that AI has negatively impacted female workers, with female workers being more at risk than their male counterparts, of unemployment following computerization. Given the rate at which AI technology is replacing manual jobs, this should be a cause for concern for policymakers and politicians alike (Acemoglu and Restrepo, 2019), particularly in terms of the potential negative impact of AI gender equality (Hamaguchi and Kondo, 2018). AI technologies are more likely to replace female workers because, globally, female workers occupy roles that are susceptible to computerization. In contrast, male workers hold decision-making positions that are unlikely to be computerized (World Economic Forum, 2019).
AI could help achieve many sustainability goals, with sustainable projects being launched based on AI algorithms and managed through AI technology. The role of AI in achieving gender equality is, however, not so clear-cut. Given that AI technology needs thousands of data points to train the AI, and the fact that there is a gender bias in data collection, AI models and developments will be skewed against women (Feast, 2019). Gender bias in AI is currently being researched to determine how to remove this gender bias so that the results from AI are more gender-equal (World Economic Forum, 2020). Such is the problem that Koyabe (2019) predicts that by 2022, 85% of AI projects will deliver false results due to the inherent gender biases in AI algorithms.
SDG 10: Reduce Inequality
AI is already being used in solving problems associated with socioeconomic inequality, including monitoring environmental quality, as environmental degradation leads to a deepening of economic inequality. Moreover, it is used to minimize food shortages because food scarcity and poor food quality lead to societal and health disparities (Stiglitz & Greenwald, 2015). Other potential applications of AI to reduce disparities include using big data techniques to improve accessibility to education, for example, matching candidates with proper instruction. The rise in AI technology and the increase in inequality offer ample opportunities to use AI technology to herald positive social change. The use of AI provides opportunities to reshape the contours of inequality because the redistribution of wealth that AI will cause will increase the overall standard of living for all (Korinek, 2017).
SDG 16: Peace, Justice and strong institutions
Regarding SDG 16, Vinuesa et al. (2020) note that strong institutions are necessary to regulate the future development of AI, and the incorporation of AI in society. There is little known about how AI could impact institutions. Positive impacts of AI in institutions include using AI algorithms to improve the detection of fraud (Vinuesa et al., 2020) and using AI technology to assess the potential effects of AI on regulations (Gorr et al., 2007). As Vinuesa et al. (2020) discuss, however, there are concerns about privacy and security regarding the incorporation of AI technology in institutions, meaning that it is unlikely that such technology will be incorporated into institutions any time soon.
Additionally, there are concerns about how algorithm bias against ethnic minorities might impact these communities, meaning that rolling out AI technology within institutions is unlikely (Ferguson, 2017). Other academics argue that using AI algorithms would expose the biases inherent within institutions, allowing the identification and correction of institutional sources of bias (Morgan, 2014). There are also concerns regarding the transparency and accountability of AI technology applied to/within institutions and a need to decide which ethical standards should be used in applied AI cases within institutions (European Commission, 2019). Data-driven AI technology is also problematic in terms of using citizen-generated data given the potential for this technology to influence opinions regarding specific products or certain political parties, for example (Harari, 2018).
It has been shown that ethical AI can potentially contribute to increasing sustainability for several of the SDGs discussed. The issue of gender equality was found to be somewhat problematic, since most of the jobs AI will eventually computerise are woman occupied. Ethical AI serves all of the SDGs discussed, in some way, with the potential impacts on sustainability depending on how AI is rolled out and how well the internal biases inherent in AI algorithms are dealt with.
AI at Venturenomix
We are working with market-leading companies in the AI domain whose models have many use cases spanning multiple sectors. We are also utilising AI in our work to support our research and project development activities. Get in touch to find out more.
Mareike Chalkley
Complemented by my 15+ years experience in founding, building and managing R&D funding consultancies in the UK and Germany, I take much personal satisfaction from supporting R&D projects that can achieve tangible, positive impact for stakeholders across social, economic and ecologic dimensions.
Get in touch
