This study explores the intersection of mobility, climate change adaptation and artificial intelligence (AI) in South Africa. As climate change impacts mobility patterns, South Africa needs effective adaptation strategies, especially to manage the mobility of internally displaced persons (IDPs) and prevent maladaptation. Artificial intelligence is used to improve lives and livelihoods in such situations, but it raises legal and policy questions. The objective of the article is to uncover the legal and policy implications of AI’s role as a climate technology for adaptation amongst IDPs in South Africa. The article utilises a qualitative desk-based approach to analyse legal frameworks, literature, official records such as parliamentary transcripts, non-governmental organisations (NGO) reports and online repositories on IDPs’ mobility patterns and adaptation, as well as the South African government’s technological interventions to support better adaptation. The findings reveal that AI can improve the planning of IDPs’ relocation. Its tools can improve needs assessment, resource allocation, protection services or infrastructure planning. However, South Africa’s legal frameworks lack specific provisions on AI. Hence, regulations are needed to address AI’s evolving role in climate adaptation in South Africa.
The study concluded that with AI policies and laws, maladaptation associated with mobility can be mitigated in South Africa.
According to the Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report:
[
Mobility is often used as a coping or adaptation strategy to escape extreme climate impact; however, some mobility typifies and results in maladaptation (Zickgraf
According to the 2022 IPCC report – Impacts, Adaptation and Vulnerability – climate change impacts may broadly be divided into two categories: the sudden or rapid onset of events (floods, storms, heatwaves and wildfires) and more gradual changes or slow onset events (coastal erosion or drought, sea level rise, desertification or ocean acidification). The Report further distinguishes the classification by demonstrating that, for instance, slow-onset events, such as ecological droughts, increase the risk of wildfire, which can lead to human displacement, loss of social ties, a sense of peace and cultural identity, as well as cause migration. Ian Fry has stated that climate mobility encompasses a range of mobility types, such as forced displacement or involuntary immobility, as well as potentially positive types, including autonomous, high agency migration and planned community-led relocation, which can be a good form of adaptation strategy (Fry
In South Africa, climate-related disruptions to human livelihoods have exacerbated the mobility of IDPs, disproportionately affecting vulnerable populations (Matikinca, Nyamakura & Shackleton
Measures, for example, aimed at assisting the IDPs to cope with displacement associated with climate sudden and slow onsets in South Africa sometimes expose them to rape, extortion, internal conflict for resources and overcrowding (Schipper
Maladaptation may include infrastructural maladaptation (Adeola
[
This means that such planning should be undertaken from the outset, when the need for relocation is established, through to the eventual integration into the new community or reintegration into the modified previous habitat. In this context, our aim is to uncover the legal implications of AI’s role as a climate technology for adaptation amongst IDPs in South Africa. Consequently, the study attempts to answer the following questions:
The article uses a qualitative desk-based approach, relying on official records such as parliamentary transcripts, Non-Governmental Organisation (NGO) reports and online repositories on IDPs’ mobility patterns and adaptation, as well as the South African government’s technological interventions to support better adaptation. As such, providing practical and legal solutions to the plight of IDPs in South Africa.
According to the IDMC, between 2008 and 2024, 232 000 people were displaced in South Africa due to 70 sudden-onset events, including flooding, wildfires and storms (IDMC
South Africa’s adaptation policy response to the climate crisis is its ‘National Climate Change Adaptation Strategy (NCCAS)’ – a comprehensive plan that outlines priority areas for addressing climate change impacts, through initiatives like early warning systems (EWS) and community awareness campaigns (The Presidency
The current adaptation initiatives have exposed gaps contextualised within the purview of disaster risk management and technological tools for climate adaptation.
The South African government has, over the years, taken steps to address the plight of climate IDPs; these have been supported by NGOs, yet there are gaps in these adaptation strategies and disaster risk management that require urgent attention. In line with these concerns, the United Nations Environment Programme (UNEP) identified gaps in the form of inadequate investment and planning on climate adaptation (UNEP
An overview of the three-phase government intervention revealed that IDPs were sheltered in community halls and provided with mattresses, food and hygiene packs (SAnews
In light of the plight of climate IDPs in South Africa, we argue that the climate change adaptation strategy and the
The IPCC special report on global warming (Hoegh-Guldberg et al.
Re-affirming the technology gaps for climate EWS and adaptation, the UN Secretary-General, António Guterres, in 2022, during COP 27, called for a global effort to ensure that the four pillars for an effective climate EWS, contained in the EW4All on Earth initiative, are implemented by different organisations (United Nations Disaster Risk Reduction [UNDRR]
Whilst tasked with the mandate to implement Pillar 1 of the EW4ALL initiative, the UNDRR uncovered significant gaps in terms of risk information and assessments worldwide, in that ‘less than half of the countries with existing EWS have access to appropriate disaster risk information, and even fewer have national legislation and regulatory frameworks for emergency response’ (UNDRR
Pillar 2, coordinated by the World Meteorological Organisation (WMO) and other cohorts, aims to shed light on the evolving nature of climate change so that governments, communities and individuals can be prepared to respond appropriately to climate crises (WMO
Pillar 3 focuses on the systematic warning strategy and communication of climate risk to individuals and communities. This initiative was led by the International Telecommunication Union (ITU) (United Nations
In South Africa, SAWS mainly uses radio and television to disseminate warnings of impending climate hazards. (SAWS
Led by the International Federation of Red Cross and Red Crescent Societies (IFRC), Pillar 4 evaluates how governments, non-governmental agencies, communities and individuals undertake anticipatory responses to climate crises (IFRC
In summary, juxtaposing South Africa’s existing adaptation and risk-reduction technology strategies with the findings and recommendations of the four Pillars of the EW4LL initiative indicates that South Africa does have in operation a combination of Pillars 1, 2 and 3. This assertion is based on the country’s warning systems, which can monitor and predict climate hazards, communicate their impacts and disseminate them to communities and the country timeously. These warning systems, although available, are not viable for early planning, evacuation or relocation of IDPs. For example, other EWS, such as the FuXi-Subseasonal model, has the capacity to significantly increase the prediction period for extreme weather from 30 days to 36 days; however, predicting potential climate disaster events as early as possible, and gaining more time for response and mitigation measures (Chen et al.
Early warnings save lives and provide vast economic benefits, with just 24 hours’ notice of an impending hazardous event, having the potential to cut the ensuing damage by 30 per cent. (UNDRR
The Global Commission on Adaptation found that ‘spending just $800 million on such systems in developing countries would avoid losses of $3–16 billion per annum’ (United Nations
South Africa’s approach to addressing the climate crisis lacks Pillar 4, representing a significant gap in the climate adaptation and risk management strategies. As alluded to by the IFCR, although the government and NGOs extend humanitarian assistance to support individuals and communities affected by disasters, the support would be more effective if well-planned before a climate hazard. The pains and neglect individuals and communities have to go through whilst in transit and trying to adapt to their new environment, according to the IFCR, can be mitigated by having the proper early warning climate technology to address climate change and associated crises (IFRC
As indicated, the IPCC AR5, IPCC {SR15} and IPCC Special Report on Climate Change and Land (SRCCL) (Olsson et al.
Also, as an adaptation tool, AI can improve shelter planning for short- and long-term adaptation strategies. AI tools, such as AI-driven smart-home systems (AIDSHs), can reduce energy usage by adjusting home appliances and monitoring and controlling lighting, temperature, security and air conditioners to adapt to extreme weather (Dąbrowska
In this context, AI systems can aid in IDPs’ effective adaptation and resilience in a new environment through AI EWS and climate AIDSHs. As noted earlier, there are growing concerns about the government’s ability to build accommodation for IDPs. This presents a good opportunity for the government to develop an innovative AI home community, where IDPs affected by climate change will be housed rather than in churches, township halls, or squatter camps. The AI intelligent-home systems, integrated with other technological systems such as biogas, can effectively manage human waste, which is then converted to energy for cooking and warmth during winter. Artificial intelligence inculcated in building plans can also help regulate temperature, monitor unsolicited activities and pre-empt heinous crimes such as rape and drug peddling, which are issues of great concern to the IDPs. Artificial intelligence systems can also help continue the education of children of climate change IDPs at the allocated centres through video and interactive learning.
Recently, scientists have unveiled climate AI technologies, such as sub-seasonal forecast tools (UNCC
AI-driven smart-home systems can be used to plan IDP shelters and accommodations to aid in better climate adaptation. The AIDSH system may be used as a smart electricity and water grid to manage heat, lighting and water appliances, conserving energy and thereby facilitating resilient adaptation (Khaleel et al.
This article is not oblivious to the legal implications of deploying AI as an adaptation tool, such as bias, invasion of privacy, security issues and the potential to contribute to greenhouse gas (GHG) emissions (UNFCCC
As AI technologies continue to evolve and permeate human lives and livelihoods, arguably, AI can be used to aid climate adaptation; however, its development and deployment must be done responsibly. None of the existing international or domestic instruments has specifically referenced AI as a climate adaptation technology. However, there are evident provisions that indirectly support the use of AI to aid climate adaptation.
The UNGP is an international instrument that identifies the rights and protection of IDPs. It also aids the IDPs in all mobility, relocation and social reintegration phases; although the UNGPs are not binding instruments, they reflect and are consistent with international human rights, such as civil and socioeconomic rights. It may also be argued that the UNGP currently constitutes customary international law (Lwabukuna
The UNGP makes no specific reference to AI as a potential technology to achieve its objectives; nevertheless, in principle, it can be argued that the UNGP supports the use of AI. This is because, for example, one of the primary uses of AI in the context of sudden-onset climate events is to facilitate climate adaptation and enhance protection for IDPs. In the context of the UNGP, AI can be deployed in ways that aid the actualisation of IDP rights.
For example, Principle 7 connotes that:
Before any decision requiring the displacement of persons, the authorities concerned shall ensure that all feasible alternatives are explored to avoid displacement altogether. Where no other options exist, all measures shall be taken to minimise displacement and its adverse effects.
The authorities undertake that such a displacement shall ensure, to the greatest practicable extent, that proper accommodation is provided for the displaced persons, that such displacements are effected in satisfactory conditions of safety, nutrition, health and hygiene, and that members of the same family are not separated.
In context, Principle 7 holds that state institutions and authorities explore feasible alternatives to avoid displacement, which might never have occurred if the right technology had been deployed. As such, the UNGP suggests, in its totality, the use of the latest technologies, such as AI EWS and AI tools, to aid climate mobility and the resettlement of IDPs. As a result, it is suggested that AI–AI-informed climate mobility initiatives and laws in South Africa should prioritise the protection of IDPs’ right to life, health and medical care; adequate food, shelter, security, dignity, violence, as well as protection from any form of discrimination and exploitation using AI technologies.
The Kampala Convention (African Union
Article IV refers explicitly to EWS technology to mitigate the impact of climate change. Article IV provides that:
The States Parties shall devise early warning systems, in the context of the continental early warning system, in areas of potential displacement, establish and implement disaster risk reduction strategies, emergency and disaster preparedness and management measures and, where necessary, provide immediate protection and assistance to internally displaced persons.
In other words, scientific evidence suggests that AI technology can be used as EWS and aid preparedness and management of the climate crisis (UNFCCC
Socioeconomic rights are the second generation of rights, which originated from the International Covenant on Economic, Social and Cultural Rights (ICESCR). Articles 11, 12 and 13 of the ICESCR recognise everyone’s fundamental rights, such as the rights to housing, health, and education, respectively, and obligate the state to take appropriate steps to achieve these rights. These rights are adopted and enshrined in the South African Constitution (SA Constitution 1996). The purpose of engraving these rights in the SA constitution is to mandate the government to formulate policies towards the actualisation of socioeconomic rights, which will equip courts to intervene when such rights are infringed upon or implemented unsatisfactorily. In theory, these rights allow citizens of signatory nations, climate change IDPs included, to
The ICESCR does not specifically refer to AI technology in its provision, although, it can be deduced that the provisions are intended to imply also the use of AI technology to actualise the rights of IDPs. Article 15(1), for example, provides that state parties to the present Covenant are to recognise the right of everyone to enjoy the benefits of scientific progress and its applications. The ICESCR encourages the use of science, which arguably includes AI technologies, to achieve the full realisation of socioeconomic rights. The Committee on Economic, Social and Cultural Rights (CESCR) General Comment No. 25, (United Nations
Moving to domestic instruments, the SA constitution is one of the most profound and progressive constitutions in the world, being that it is founded on dignity, equality and freedom. These rights are also extended to IDPs as they are citizens of South Africa. The government, under the constitution, is obliged to respect and protect the dignity of IDPs by providing them with the right to proper housing, education and a clean and healthy environment. Sections 24, 26, 27 and 29 of the SA
‘Right to an environment that is not harmful to their well-being and to have it protected for the benefit of present and future generations’.
‘Right to have access to adequate housing, which the government must take reasonable legislative and other measures within its available resources to achieve the progressive realisation of the right’.
‘Right to health care, water and social security’.
‘Right to education’.
The inclusion of the rights to health, education and housing in the constitution, however, often does not translate into action, especially in the context of IDPs; hence, it is argued that the government should provide the IDPs with basic rights, by developing AI systems that support climate change adaptation efforts, that will aid the actualisation of the above-mentioned rights. The South African government is a signatory to the socio-economic rights and is bound to fulfil measures that will enable IDPs to enjoy these rights, including using technology such as AI to promote environmental education, security and access to basic health care and information. The CESCR’s General Comment No. 25 (United Nations
The primary aim of the DMA (Republic of South Africa
It is worth noting that, as previously discussed, the entire sub-paragraphs of Section 27 of the DMA are relevant, yet the government has not taken the necessary measures to address mobility and disaster management issues affecting the IDPs. Section 27(2)(m) may be read as utilising AI EWS to facilitate responses to the climate crisis and to support disaster risk management for post-disaster recovery and rehabilitation. As such, Section 27(2)(n) should be read as taking steps, such as integrating AI into other technologies like smart-home systems and biogas technology; taking such steps is crucial to prevent maladaptation. Finally, Section 27(2)(o) can be construed as requiring the Minister of Cooperative Governance and Traditional Affairs (CoGTA), in cooperation with other cabinet members or climate change experts, to seek ways to facilitate international assistance/finance for the development and deployment of AI for climate mitigation and adaptation. The CoGTA and other relevant stakeholders, in addition to sourcing funds locally, should take a bold step to engage those historically responsible for the current climate change and its consequences, as well as to facilitate the adoption of AI as a climate mobility and adaptation technology in South Africa. In summary, this article suggests integrating AI into existing laws, particularly the DMA.
The NCCAS makes no specific reference to AI in its plan to aid climate adaptation; however, the framework suggests strategic intervention to promote and invest in research on the most effective adaptation responses to different climate change impacts (Republic of South Africa
It should, however, be noted that, although the NCCAS was developed to offer policy guidelines for the climate change crisis, the strategy itself was not associated with a specific climate change policy; hence, the crafting of a specific national climate change policy, which will be discussed in the ensuing Section.
The NCCR policy provides a foundation for South Africa’s climate change response, prioritising sustainable development, economic growth and human well-being (Republic of South Africa
The NCCR policy also lacks a clear AI regulatory framework for climate change, let alone for those displaced by the climate crisis. It nevertheless encourages the use of technology, such as EWS, to address the climate change crisis (Garland
In general, Paragraph 11 of the NCCR policy, read with Sections 24, 26, 27 and 29 of the SA constitution and Article 11, 12 and 13 of the ICESCR, requires that states, in this context, the South African government, to progressively make provisions for the actualisation of socioeconomic rights of its citizens, including climate change IDPs. This view is further strengthened by international frameworks such as the Kampala Convention, UNGP and the ICESCR, which address the plight of climate change IDPs. Section 233 of the SA
When interpreting any legislation, every court must prefer any reasonable interpretation of the legislation that is consistent with international law over any alternative interpretation that is inconsistent with international law.
An overview of Section 233 read with Section 39 of the SA
The interpretation of Section 27 (1)(2) of the DMA and the NCCR policy must be consistent with provisions, principles and guidelines as prescribed in international laws, which proffer that AI technologies can be used to mitigate and aid adaptation to climate change in South Africa. It also advocates amending the NCCR policy to promote innovation and the deployment of AI climate technology tools, whilst ensuring safety, fairness and transparency in their development and deployment.
Climate change poses a significant threat in South Africa, including increased displacement of the affected population. Whilst most IDPs use mobility as a form of coping strategy to escape the climate crisis, there is a need to ensure that such mobility does not translate to other forms of maladaptation. There is some evidence that AI could make significant additional contributions to adaptation strategies in the years ahead. Artificial intelligence is not the ultimate panacea for climate change; however, studies have shown that AI EWS, compared to the conventional EWS, can better predict climate-change crises, enabling proactive planning, evacuation and relocation of IDPs. In addition, AI can aid better adaptation regarding humanitarian support and infrastructural design. That said, integrating AI with other climate technologies raises both legal and policy considerations, given AI’s dual role as both a potential climate adaptation technology and a potential environmental threat. As a result, regulatory frameworks must be put in place to guide the development and deployment of AI as a climate mobility and adaptation technology. This article, hence, recommends using AI for climate mobility and adaptation in South Africa.
This group will include experts from private and state institutions, academia and the AI community, who will guide AI deployment and regulations. Currently, the government has a series of forums – parliament, the Inter-Ministerial Committee on Climate Change (IMCCC), civil societies, the Forum of South African Directors-General clusters, the Intergovernmental Committee on Climate Change (IGCCC) and the National Disaster Management Council – which can be deployed for this task. Any of these groups or cohorts can guide the development of AI climate policies to regulate the deployment of AIDSHs and EWS; these policies will address issues of liability and accountability, human rights concerns and deployment as a climate adaptation tool. The team, amongst its duties, should be able to define AI terminologies and concepts and set safety and security standards for deploying and using AI. In collaboration with the government, the team should mobilise the resources necessary for climate mitigation and adaptation. This may be in the form of financial resources, technical cooperation and technology transfers at domestic, sub-regional, regional and international levels.
It is paramount that the AI task team engage with civil society, private and public institutions, experts and international institutions to gather input on AI regulations from diverse perspectives.
We recommend that SAWs upgrade existing warning systems and collaborate with other stakeholders to increase response time for mobility planning and adaptation. Although SAWs have notably disseminated early warning information via radio and television stations, it would be more worthwhile to also use SMSs, as recommended by the ITU.
Finally, it is further recommended that Section 27 of the DMA and Paragraph 11 of the NCCR policy be read to imply that the South African government shall use or consider the use of AI EWS and AIDSHs to aid mobility and adaptation/resettlement of IDPs in their new environment or housing, rather than in community halls, schools, churches, shacks, squatter camps and other dehumanising environments. As a matter of policy, it would be prudent to set aside resources for long-term planning for climate mobility and the adaptation/resettlement of IDPs, as contemplated in the NCCR policy. As promised, the South African government should fulfil its commitment to progressively build the 4,396 temporary accommodations promised to KZN’s IDPs in Ilembe District, Ugu and eThekwini. Such accommodation should reflect the smart-home systems, integrated with other technological systems, such as biogas.
A preprint version of this work was previously published on ResearchGate and EarthArXiv. Available at DOI:10.31223/X5D729.
The authors reported that they received funding from the CGIAR-MMUST Grant, Africa Climate Mobility Academy, which may be affected by the research reported in the enclosed publication. The author has disclosed those interests fully and has implemented an approved plan for managing any potential conflicts arising from their involvement. The terms of these funding arrangements have been reviewed and approved by the affiliated University in accordance with its policy on objectivity in research.
Temitope A. Obisanya: Conceptualisation, Formal analysis, Funding acquisition, Methodology, Writing – original draft and Writing – review & editing. Ademola O. Jegede: Conceptualisation, Funding acquisition, Methodology, Supervision, Writing – original draft and Writing – review & editing. All authors reviewed the article, contributed to the discussion of results, approved the final version for submission and publication, and take responsibility for the integrity of its findings.
This article followed all ethical standards for research without direct contact with human or animal subjects.
Data sharing is not applicable to this article as no new data were created or analysed in this study.
The views and opinions expressed in this article are those of the authors and are the product of professional research. They do not necessarily reflect the official policy or position of any affiliated institution, funder, agency or that of the publisher. The authors are responsible for this article’s results, findings and content.