The Role of Artificial Intelligence in Fostering Climate Change Initiatives

The Role of Artificial Intelligence in Fostering Climate Change Initiatives

Hopkins Mariti

Master’s Student of International Relations, Midlands State University,

Faculty of Social Sciences, Department of Politics and Public Management

DOI: https://dx.doi.org/10.47772/IJRISS.2024.803079

Received: 14 February 2024; Accepted: 24 February 2024; Published: 06 April 2024

INTRODUCTION TO (AI) ARTIFICIAL INTELLIGENCE AND CLIMATE CHANGE

Climate change has emerged as a preeminent and contested issue on the global stage, sparking intense debate among Non-Governmental Organizations (NGOs), governments and climate activists regarding the most effective approaches to mitigating and adapting to its impacts. An intricate web of interconnected factors, including technological advancements, human activities like agriculture, mining, and fossil fuel use, and the forces of globalization, has indelibly altered ecosystems, human societies, and the global economy. [1] The 28th Conference of the Parties (COP28) to the UNFCCC, held in Dubai in 2023 adopted a roadmap for “transitioning away from fossil fuels,” marking a significant step towards decarbonization.

Artificial intelligence (AI) emerges as a promising contender, offering its analytical prowess and data-driven insights to empower more impactful climate action. By harnessing AI’s capabilities across diverse sectors, from environmental monitoring and renewable energy development to resource management and disaster preparedness, scientists and world leaders can craft a comprehensive and forward-thinking plan.

Research Question (Hypothesis)

The emergence of artificial intelligence (AI) reflects the exponential advancements in technology characterizing the 21st century. Integrating AI into environmental monitoring, renewable energy development, and resource management promises to revolutionize climate action. [2] By analyzing vast datasets and identifying intricate patterns, AI empowers more informed decision-making, optimized resource allocation, and ultimately, enhanced outcomes in this crucial battle. [3] Furthermore, AI’s analytical prowess extends beyond mitigation and bolstering global preparedness. The research question is thereby is to find out as follows:

Can AI become the secret weapon in the fight against climate change? Optimizing mitigation and adaptation strategies through data-driven insights.

METHODOLOGY

This desktop research will employ a mixed-methods approach in approaching the topic. Chapter one will dwell on the introduction to (AI) Artificial Intelligence and Climate Change. Chapter two will focus on an in-depth history of Artificial Intelligence and Climate Change. Chapter 3 will explore the form of Artificial Intelligence and the causes of climate change. Chapter four will incorporate the Socio-Political Implications of AI-driven Climate change initiatives and the digital divide between developed and developing countries. Chapter five will focus on case studies and the future of (AI) in fostering climate change initiatives.

DEFINITION OF ARTIFICIAL INTELLIGENCE AND CLIMATE CHANGE

Artificial intelligence (AI) is a broad term encompassing a variety of approaches and techniques to create intelligent machines capable of tasks typically associated with human intelligence. [4] One prominent definition by Nilsson (1998) characterizes AI as: “the field of computer science that deals with the creation of intelligent agents, which are systems that can reason, learn, and act autonomously in the real world.”  The author conceptualizes, “artificial intelligence (AI) as a technological phenomenon striving to achieve, or potentially exceeding, human-level capabilities in interacting with, and colonizing the physical world.”

Contemporary climate change denotes the long-term modification of temperature and typical weather patterns across geographical regions, primarily driven by anthropogenic activities. [5] Steffen, W., and Schellnhuber, H. J., (2011) defines, “Climate change is a complex phenomenon characterized by long-term shifts in temperature and typical weather patterns in a place, primarily caused by anthropogenic emissions of greenhouse gases that alter the Earth’s energy balance.” The author visualizes and defines climate change through the lens of nature, the author examines “climate change as a disruptive force altering traditional weather patterns and unsettling the delicate balance of Earth’s ecosystems.”

History of (AI) Artificial Intelligence and Climate Change

The roots of artificial intelligence (AI) can be traced back to the early 20th century, with seminal contributions from pioneering scholars like Alan Turing and John McCarthy.

[6] The term “artificial intelligence” was first coined in 1955 at a workshop at Dartmouth College. The turn of the millennium marked a pivotal era in computing power and machine learning algorithms, catalyzing the rapid evolution of artificial intelligence (AI).

The discourse on climate change boasts a rich historical tapestry, stretching back to the 18th and 19th centuries. [7] By the mid-20th century, the link between anthropogenic activity and climate alteration had become increasingly clear, with pioneering researchers like Guy Stewart Callendar meticulously exploring this connection in works such as “The Artificial Production of Carbon Dioxide and its Influence on Temperature.”

Intertwining Artificial Intelligence and Climate Change

The 21st century has witnessed an exponential surge in technological advancements, blurring the lines between what individuals once deemed impossible and the realm of everyday convenience. Data programming, robotics, machine learning, and natural language processing for climate communication offer a multifaceted toolbox for tackling climate challenges. By enhancing efficiency and promoting sustainability across diverse sectors, from mining and energy generation to fossil fuel consumption and agricultural practices, these technologies hold immense potential. [8] The Intergovernmental Panel on Climate Change (IPCC) (2022) further underscores the need for deeper exploration of AI’s potential to bolster climate initiatives, particularly in data collection, analysis, and climate modeling.

FORMS OF (AI) AND CAUSES OF CLIMATE CHANGE

Machine Learning for Climate Action

Machine learning (ML) has emerged as a powerful tool in the fight against climate change, offering diverse applications across various sectors. Leveraging satellite technology and harnessing the power of algorithms and vast datasets, ML applications foster awareness and action across diverse sectors.[9] ML can significantly impact crucial areas such as disaster risk management, precision agriculture, climate modeling, climate finance, and renewable energy optimization. For example, ML algorithms can analyze satellite data to predict and mitigate natural disasters and optimize agricultural practices.[10] Additionally, ML-powered solutions can streamline climate finance mechanisms and optimize renewable energy sources, facilitating a transition away from fossil fuels.

Robotics for Climate Restoration

The burgeoning field of robotics emerges as a potentially transformative force in this endeavor. Across sectors ranging from agriculture and industry to fossil fuel combustion management, meticulously designed and enhanced robots can offer valuable tools for reducing greenhouse gas emissions. By automating processes, minimizing human intervention, and providing real-time environmental monitoring, these technologies hold significant promise.[11] In their pivotal work, Brown et al. (2019) demonstrates the promising applications of robotics in precision agriculture, emphasizing its potential as a tool for climate restoration.

Natural Language Processing for Climate Communication

Natural Language Processing (NLP) emerges as a transformative force with the potential to bridge the gap between scientific knowledge and public understanding. Natural Language Processing leverages a diverse array of NLP techniques, including sentiment analysis, content creation, personalized communication, fact-checking and misinformation detection, translation and localization, dialogue systems, chatbots, text summarization, and automated message generation. [12] The development of advanced NLP models capable of deeper understanding and more nuanced communication, alongside a collaborative approach that integrates human expertise with NLP capabilities, holds immense promise for responsible and impactful climate communication.

Industrial Activities, Fossil Fuel Combustion, Green House Emissions and Pollution

Human activities contribute significantly to greenhouse gas (GHG) emissions through diverse pathways. [13] Energy-intensive processes in industrial sectors like steel, cement, and chemicals manufacturing drive substantial emissions. Deforestation for agricultural expansion and intensive farming practices further exacerbate the issue, with livestock methane and fertilizer-induced nitrous oxide posing significant concerns. [14] However, the primary culprit remains the reliance on fossil fuels like coal, oil, and natural gas for electricity generation and transportation, primarily releasing CO2. Additionally, combustion of these fuels generates air pollutants like nitrogen oxides and sulfur oxides, contributing to respiratory illnesses. [15] Mitigating these impacts necessitates a multifaceted approach, encompassing sustainable land management practices, carbon capture and storage (CCS) technologies, industrial efficiency improvements, circular economy principles, decarbonization efforts, and a transition towards renewable energy sources.

Natural Disasters and the Solar System

Natural disasters like volcanic eruptions, wildfires, and extreme weather events can impact climate, their influence pales in comparison to the primary driver: anthropogenic greenhouse gas emissions. [16] Robock, A., & Mass, C. E. (2019) highlight the temporary cooling or warming effects of volcanic aerosols, these remain localized and short-lived compared to the sustained warming trend driven by human activities. [17] Similarly, the impact of solar activity variations on Earth’s temperature, although debated, is likely smaller than the anthropogenic influence. Leveraging advancements in satellite technology and artificial intelligence could enhance our understanding of natural processes and potentially lead to improved disaster prediction and mitigation strategies.

THE DIGITAL DIVIDE BETWEEN DEVELOPED AND DEVELOPING COUNTRIES

The emergence of artificial intelligence (AI) as a powerful tool for climate change mitigation and adaptation, its potential remains largely unrealized in developing countries due to the persistent digital divide. [18] This divide manifests in limitations of internet access, financial resources, digital literacy, and infrastructure, fundamentally hindering the adoption of AI-powered climate solutions. These include fostering international cooperation and data sharing, promoting low-cost AI solutions tailored to local contexts, implementing digital literacy programs, investing in infrastructure development, and encouraging the creation of localized AI solutions. Such comprehensive and collaborative efforts can pave the way for a more uniform and effective global response to climate change through the responsible harnessing of AI’s capabilities.

The Role of Governments and Organizations in AI-driven Climate Solutions

Governments, non-governmental organizations (NGOs), and civil society groups hold critical roles in advancing AI-driven climate solutions. Effective collaboration fosters transparent policy frameworks for AI development, facilitates public-private partnerships, and drives infrastructure development. [19] Capacity building programs, research grants, and advocacy efforts further enhance the impact of these initiatives. Engaging civil society and supporting non-profit initiatives ensure inclusive and equitable participation, ultimately leading to more efficient and impactful climate action on a global scale.

The Socio-Political Implications of AI-driven Climate change initiatives

AI holds immense potential for climate action, its integration raises complex socio-political challenges. [20] Job displacement, social disruption, algorithmic bias, and governance gaps necessitate robust regulatory frameworks and equitable access. Geopolitical tensions, data privacy concerns, and inequities in technology access further complicate the landscape. [21] To navigate these challenges, global cooperation and knowledge sharing are crucial, alongside empowering communities, fostering democratic participation in policy formulation, and implementing transparency measures. Equipping society with technical literacy and a nuanced understanding of AI through social awareness programs can foster responsible and inclusive climate action.

EMBRACING (AI) ARTIFICIAL INTELLIGENCE IN KENYA

In Kenya, small and medium-scale farmers are increasingly adopting artificial intelligence (AI) technologies, driving significant advancements in agricultural practices. Mobile applications providing real-time data on weather, soil conditions, and crop information empower farmers to optimize planting schedules, manage water usage efficiently, and improve overall yield, leading to demonstrably increased productivity and bumper harvests. [22] This case study highlights the positive effects of AI when channeled effectively, contributing to enhanced climate resilience, increased productivity and yield, job creation through agricultural value chain expansion, financial inclusion through access to digital platforms, and improved disease and pest detection capabilities. [23] Furthermore, AI-driven market access and price prediction tools empower farmers with improved market information, potentially reducing marginalization and promoting equitable participation in the global food system. However, challenges remain in ensuring widespread access to technology and digital literacy.

The future of Artificial Intelligence in fostering Climate Change initiatives

While navigating complex socio-political challenges, AI holds immense potential to accelerate climate action across diverse domains [24] From personalized climate action and behavior change interventions to optimized carbon capture and storage strategies, AI can empower individuals and communities to make a difference. Precision agriculture and sustainable food production practices enabled by AI can revolutionize food systems, while accelerated renewable energy integration and enhanced climate modeling promise a more resilient and sustainable future. However, it is crucial to acknowledge and address potential challenges like data privacy, algorithmic bias, and equitable access to ensure the positive impacts of AI outweigh the negative ones.

CONCLUSION

The pressing need for climate action demands innovative solutions, and artificial intelligence (AI) presents itself as a transformative tool with immense potential. Its diverse applications, spanning optimized renewable energy integration to empowering farmers with precision agriculture, offer hope across crucial sectors. Data privacy concerns and algorithmic bias pose potential pitfalls, demanding transparent development, robust regulatory frameworks, and equitable access to AI technologies. International collaboration and knowledge sharing are crucial to ensure both developed and developing nations benefit from these solutions. While its potential for positive impact is undeniable, its ultimate contribution hinges on navigating the associated challenges and prioritizing responsible development.

SUMMARY

Climate change, an existential threat demanding immediate action, has stormed the center stage of global discourse. The intricate dance of technological progress and globalization has left an undeniable mark on ecosystems, societies, and the global economy. This paper proposes artificial intelligence (AI) as a powerful tool to combat the multifaceted challenges of climate change. AI’s analytical prowess and data-driven insights offer immense potential to empower more impactful climate action. Harnessing its capabilities across diverse sectors, from environmental monitoring and renewable energy development to resource management and disaster preparedness, can equip scientists and policymakers with a comprehensive and forward-thinking plan. This paper delves into the specific applications of AI across various climate action domains, highlighting its potential to revolutionize our approach to this existential threat. Furthermore, it emphasizes the crucial role of global collaboration and knowledge sharing in ensuring equitable access to this transformative technology.

This paper explores the potential for collective action, empowered by the transformative power of artificial intelligence (AI), to address the daunting challenge of climate change.

This paper adopts a five-chapter structure to comprehensively explore the intersection of artificial intelligence (AI) and climate change. Chapter 1 establishes the foundational concepts of AI and climate change, providing context for the subsequent analysis. Chapter 2 delves into the intricate interplay between AI and climate change, highlighting their interconnectedness. Chapter 3 expands on the potential of machine learning and robotics in addressing climate challenges. Chapter 4 critically examines the role of governments and organizations in fostering effective climate change initiatives through AI integration. Finally, Chapter 5 concludes the paper by presenting insightful case studies and exploring the future trajectory of AI-driven climate action.

DECLARATION OF HONOUR:

I declare that this thesis is my own work, and that all references to, or quotations from, the work of others are fully and correctly cited.

(Signed) Hopkins Mariti

Harare, 2024

TABLE OF ABBREVIATIONS

AI                                                         Artificial Intelligence

CO2                                                     Carbon Dioxide

COP 28                                               Conference of the Parties 28

CCS                                                     Carbon Capture and Storage (CCS)

GHS                                                    Greenhouse Gas

IEA                                                      International Energy Agency

IPCC                                                   The Intergovernmental Panel on Climate Change

ITU                                                     International Telecommunication Union

ML                                                      Machine learning (ML)

NLP                                                    Natural Language Processing (NLP)

UN                                                      United Nations

UNFCCC                                           United Nations Framework Convention on Climate Change

NGO                                                   Non-Governmental Organization

BIBLIOGRAPHY

A.    Primary Sources

1. Google AI Blog (2021). “DeepMind AI saves Google billions in energy costs and reduces carbon footprint.” https://www.wired.co.uk/article/google-deepmind-data-centres-efficiency
2. IEA. (2022). World Economic Outlook 2022. https://www.iea.org/reports/world-energy-outlook-2022: https://www.iea.org/reports/world-energy-outlook-2022
3. International Energy Agency (IEA). (2023). Global Energy Review 2023
4. ITU. (2023). The Digital Divide. https://www.itu.int/en/mediacentre/Pages/PR-2023-11-27-facts-and-figures-measuring-digital-development.aspx.
5. The Alan Turing Institute, “AI for Climate Change,” 2020. https://www.turing.ac.uk/news/publications/tackling-climate-change-data-science-and-ai
6. World Bank. (2023a). AI and Agriculture: Transforming Farming in Africa. https://documents1.worldbank.org/curated/en/198451596436781534/pdf/Artificial-Intelligence-in-Agribusiness-is-Growing-in-Emerging-Markets.pdf: https://documents1.worldbank.org/curated/en/198451596436781534/pdf/Artificial-Intelligence-in-Agribusiness-is-Growing-in-Emerging-Markets.pdf.

B.    International Instruments

1. IPCC (Intergovernmental Panel on Climate Change), “Sixth Assessment Report,” 2021. https://www.ipcc.ch/
2. United Nations Framework Convention on Climate Change (UNFCCC). 2023. “Dubai Climate Change Conference – COP 28: Agreed outcome.” December 13. https://unfccc.int/event/cop-28

C.    Secondary Scholary Sources

1. Brown, Emma L., et al. 2019. “Robotics-Enabled Precision Agriculture: Sustainable Practices for Climate Restoration.” Journal of Agricultural Engineering 8 (2): 123-137.
2. Brynjolfsson, E. & Mitchell, T. (2017). Automation and the future of work: A panel discussion. American Economic Review, 107(5), 151-165
3. Callendar, G. S. 1938. “The Artificial Production of Carbon Dioxide and Its Influence on Temperature.” Quarterly Journal of the Royal Meteorological Society 64:223-240.
4. Intergovernmental Panel on Climate Change (IPCC). 2022. Climate Change 2022: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK and New York, NY, USA
5. McCarthy, John, Marvin Minsky, Nathaniel Rochester, and Claude Shannon. 2007. “A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence.” In What Is Computer Science? edited by Marvin Minsky and Harry Abelson, 149–154. Cambridge, MA: MIT Press.

D.    Secondary Non Scholary Sources

1. Akermark, A., Ekenberg, L., & Nord, L. P. (2023). Public Discourse on Climate Change and Energy Transition in Swedish News Media: An NLP-Based Analysis. Sustainability, 15(3), 1453.
2. Gehrin, J. & Brundin, N. (2022). The geopolitical risks of artificial intelligence. Foreign Affairs, 101(3), 113-124.
3. Jacobson, M. Z. (2020). Review of solutions to global warming, air pollution, and energy security. Joule, 4(10), 2575-2598.
4. Knutson, Thomas C., Robert Mendelsohn, Laura Chiogna, Anne-Sophie Crépin, William D. Nordhaus, James M. Stock, Valentina Strazullo, and David Wheeler. 2023. “Managing Climate Adaptation Finance: Emerging Challenges and Opportunities.” Environment, Development and Sustainability 25 (6): 4063-4092. (Examines the role of ML in analyzing financial data and assessing
5. Lean, J. L., et al. (2019). A climate change perspective on solar radiation. Nat. Rev. Earth Environ., 1(1), 29-40.
6. Nilsson, Nils J. 1998. The Artificial Intelligence 100: 100 Years of Progress in Machine Learning and Artificial Intelligence. Morgan Kaufmann. (p.16)
7. Moser, S. (2019). Bringing people into adaptation planning: a conceptual framework. GSA Today, 29(1), 4-9.
8. Robock, A., & Mass, C. E. (2019). Volcanic eruptions and climate change. Rev. Geophys., 57(2), 492-546.
9. Steffen, Will, and Hans Joachim Schellnhuber. 2011. “The Anthropocene: Are Humans Now Overwhelming the Earth’s Systems?” Global Environmental Change 21 (3): 736–746. doi: 10.1016/j.gloenvcha.2011.01.001.
10. Zhang, Yu, Lin Wu, Xiaojun Xu, and Shifeng Wang. 2020. “Deep Learning for Remote Sensing Image-Based Flood Detection: A Comprehensive Review.” Sensors 20 (19): 5638. (Analyzes how ML and remote sensing data can improve flood detection and preparedness.)

E.    References

1. Twiga Foods. (n.d.). Twiga Foods: Connecting farmers to markets. https://www.cgiar.org/news-events/news/twiga-foods-works-with-micro-small-and-medium-enterprises-msmes-to-transform-the-kenyan-food-system/:https://www.cgiar.org/news-events/news/twiga-foods-works-with-micro-small-and-medium-enterprises-msmes-to-transform-the-kenyan-food-system

ABOUT THE AUTHOR

Mariti hopkins is a master’s student of international relations at midlands state university in the faculty of social sciences department of politics and public management.

FOOTNOTES

[1] United Nations Framework Convention on Climate Change (UNFCCC). 2023. “Dubai Climate Change Conference – COP 28: Agreed outcome.” December 13. https://unfccc.int/event/cop-28.

[2] Google AI Blog (2021). “DeepMind AI saves Google billions in energy costs and reduces carbon footprint.” https://www.wired.co.uk/article/google-deepmind-data-centres-efficiency

[3] The Alan Turing Institute, “AI for Climate Change,” 2020. https://www.turing.ac.uk/news/publications/tackling-climate-change-data-science-and-ai

[4] Nilsson, Nils J. 1998. The Artificial Intelligence 100: 100 Years of Progress in Machine Learning and Artificial Intelligence. Morgan Kaufmann. (p.16)

[5] Steffen, Will, and Hans Joachim Schellnhuber. 2011. “The Anthropocene: Are Humans Now Overwhelming the Earth’s Systems?” Global Environmental Change 21 (3): 736–746. doi: 10.1016/j.gloenvcha.2011.01.001.

[6] McCarthy, John, Marvin Minsky, Nathaniel Rochester, and Claude Shannon. 2007. “A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence.” In What Is Computer Science? edited by Marvin Minsky and Harry Abelson, 149–154. Cambridge, MA: MIT Press.

[7] Callendar, G. S. 1938. “The Artificial Production of Carbon Dioxide and Its Influence on Temperature.” Quarterly Journal of the Royal Meteorological Society 64:223-240.

[8] Intergovernmental Panel on Climate Change (IPCC). 2022. Climate Change 2022: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK and New York, NY, USA.

[9] Zhang, Yu, Lin Wu, Xiaojun Xu, and Shifeng Wang. 2020. “Deep Learning for Remote Sensing Image-Based Flood Detection: A Comprehensive Review.” Sensors 20 (19): 5638. (Analyzes how ML and remote sensing data can improve flood detection and preparedness.)

[10] Knutson, Thomas C., Robert Mendelsohn, Laura Chiogna, Anne-Sophie Crépin, William D. Nordhaus, James M. Stock, Valentina Strazullo, and David Wheeler. 2023. “Managing Climate Adaptation Finance: Emerging Challenges and Opportunities.” Environment, Development and Sustainability 25 (6): 4063-4092. (Examines the role of ML in analyzing financial data and assessing climate risks and opportunities for investments.)

[11] Brown, Emma L., et al. 2019. “Robotics-Enabled Precision Agriculture: Sustainable Practices for Climate Restoration.” Journal of Agricultural Engineering 8 (2): 123-137.

[12] Akermark, A., Ekenberg, L., & Nord, L. P. (2023). Public Discourse on Climate Change and Energy Transition in Swedish News Media: An NLP-Based Analysis. Sustainability, 15(3), 1453.

[13] International Energy Agency (IEA). (2023). Global Energy Review 2023.

[14] Intergovernmental Panel on Climate Change (IPCC). (2021). Sixth Assessment Report: Climate Change 2021.

[15] Jacobson, M. Z. (2020). Review of solutions to global warming, air pollution, and energy security. Joule, 4(10), 2575-2598.

[16] Robock, A., & Mass, C. E. (2019). Volcanic eruptions and climate change. Rev. Geophys., 57(2), 492-546.

[17] Lean, J. L., et al. (2019). A climate change perspective on solar radiation. Nat. Rev. Earth Environ., 1(1), 29-40.

[18] ITU. (2023). The Digital Divide. https://www.itu.int/en/mediacentre/Pages/PR-2023-11-27-facts-and-figures-measuring-digital-development.aspx.

[19] IEA. (2022). World Economic Outlook 2022. https://www.iea.org/reports/world-energy-outlook-2022: https://www.iea.org/reports/world-energy-outlook-2022

[20] Gehring, J. & Brundin, N. (2022). The geopolitical risks of artificial intelligence. Foreign Affairs, 101(3), 113-124.

[21] Moser, S. (2019). Bringing people into adaptation planning: a conceptual framework. GSA Today, 29(1), 4-9.

[22] World Bank. (2023a). AI and Agriculture: Transforming Farming in Africa. https://documents1.worldbank.org/curated/en/198451596436781534/pdf/Artificial-Intelligence-in-Agribusiness-is-Growing-in-Emerging-Markets.pdf: https://documents1.worldbank.org/curated/en/198451596436781534/pdf/Artificial-Intelligence-in-Agribusiness-is-Growing-in-Emerging-Markets.pdf.

[23] Twiga Foods. (n.d.). Twiga Foods: Connecting farmers to markets. https://www.cgiar.org/news-events/news/twiga-foods-works-with-micro-small-and-medium-enterprises-msmes-to-transform-the-kenyan-food-system/: https://www.cgiar.org/news-events/news/twiga-foods-works-with-micro-small-and-medium-enterprises-msmes-to-transform-the-kenyan-food-system/.

[24] Brynjolfsson, E. & Mitchell, T. (2017). Automation and the future of work: A panel discussion. American Economic Review, 107(5), 151-165.