Artificial Intelligence: A blessing or a curse? What socio-economic challenges will humanity face in the next decades?
- TechnicallyMe
- 6 days ago
- 7 min read
Finalist in the Northeastern University London Essay Competition 2025
Artificial Intelligence (AI) is reshaping our world in ways once considered unimaginable. While we are beyond the era envisioned by futurist Alvin Toffler in the 1970s, the rapid integration of AI technologies still brings the same challenge: predicting what the future will hold. From healthcare to education, AI is revolutionising industries, creating vast opportunities while also presenting significant socio-economic challenges. Given these profound impacts, this essay will argue that AI is ultimately a blessing, where the risks are outweighed by the upside benefits. In particular, I will argue that the socio-economic challenges of rising inequality and climate change that humanity is facing will make AI not just a blessing but a necessity.
The global economic divide is a pressing issue, and the concern is that AI will worsen this. In 2021, not only did the top 1% have a wealth share of 45.6% (Equentis, 2022), but the wealth of the top 0.01% rose from less than 8% to almost 12% between 1995 and 2023 (Wid.World, 2023). One challenge humanity faces is the potential concentration of AI in these wealthier regions that are more likely to implement AI technologies, thus deepening global inequality. This can be seen as the USA and China’s private investment into AI were €62.5 billion, €7.3 billion respectively in 2023 whereas the next 13 nations combined only have €17.1 billion invested in total (European Parliament, 2023). This disparity in investment emphasises that wealthier and progressive nations may benefit more from the economic growth that AI brings. By “using advances in natural language processing”, AI can increase global GDP by 7% and even lead to an increase in productivity by 1.5% over the next 10 years (Goldman Sachs, 2023). Currently this is likely to be in favour of developed nations rather than developing countries, for the following key reasons:
The rate of integration and adoption of AI is going to diverge as forward-thinking countries like the USA, China, and some European nations prioritise AI integration as a national policy and make it a strategic priority to boost their industries while developing economies remain stagnant and unable to compete.
Wealthier nations would also benefit from a greater influence over the development and legislation of AI technologies. This is already occurring as countries like the USA have developed their own regulations over AI technology - the National AI Initiative Act of 2020 for example. It is therefore a concern that over time global AI regulations will prioritise developed countries at the expense of the interests of developing countries, limiting their ability to shape AI development in ways that could assist their unique needs and challenges.
In countries where AI is adopted, production costs could be reduced significantly and this would compete with production methods in less-developed economies, causing losses in the latter with substantial consequences. For instance, developing country businesses may have to lower wages or lay off workers to stay competitive, which could push more households into poverty.
Finally, a critical differentiator between nations' uptake of AI will depend on how open their public are towards it. Geert Hofstede’s 6D cultural model constitutes a relevant background against which we can see how populations relate to the continuous transformation prompted by the rapid AI development. Societies with high ‘individualism’ and lower ‘power distance’ are more open to new ideas and thus more likely to embrace emerging AI technologies. (Hofstede, 2010, 2020).
The implementation of AI has the potential to not just be a blessing but is a necessity for many developing countries to provide urgently needed solutions to their most pressing challenges - it could provide substantial benefits in the healthcare sector with a much needed impact on severe healthcare crises particularly in sub-Saharan Africa: in 2020, 58% of the population of Somalia, 51% of the population in Chad and a total of 408.6 million people within this region were left without access to healthcare (WHO, 2020-2022). With AI’s potential to increase efficiency, enhance accessibility and optimise scarce resources, it can help address these issues and lead to breakthroughs in not just healthcare but in many other sectors as well. AI-powered technologies are already improving diagnostics, cancer detection, drug discovery, and even treatment accuracy across the globe. In fact, AI has seen up to a 98.58% accuracy in detecting early stages of breast cancer with recurrent neural networks (RNN) (Darbandi et al., 2024). With the help of AI, developing economies would have the ability to combat their healthcare crises, improving the quality of life for millions, and stimulating economic growth. In doing so, these nations can begin closing the gap of healthcare standards between themselves and the more developed parts of the world, creating a healthier future for their populations.
Another major topic to address is the effect AI could have on the environment and climate change. It is worth bearing in mind that many climate change consequences are borne disproportionately by developing countries, with rising water levels, and poor agricultural yields, affecting poorer areas of the world. As AI’s processing capabilities grow, we are beginning to see its use cases in efficiency and optimization, improving energy usage across various industries. For example, in manufacturing and power generation, AI could help optimise carbon capture and storage (CCS) processes reducing energy consumption while limiting the emissions of greenhouse gases. Smart grids are also a focus point for AI optimisation, making it possible to receive a large amount of data simultaneously.
However, critics may argue that the extensive processing of AI itself has its own environmental effects too. For instance, the computing power used to back this processing, as well as to train AI models, could contribute to greater carbon emissions. For example, Google’s increased artificial intelligence usage has caused its greenhouse emissions to increase by 48% in just 5 years (The Guardian, 2024). My view is that overall environmental effects may be reduced as the future of AI is shifting towards specialised, smaller models that perform narrow, specific tasks. In comparison to general-purpose models like GPT-4, these use far less computational power and are much more energy efficient, meaning that they have a significantly lower carbon footprint. Additionally, as AI chips become more efficient, the energy required to train these models would also decrease and the current model of training AI systems may plateau. This would lead to a decrease in the amount of computational power required to achieve further performance gains, giving rise to more efficient AI training techniques that would further reduce its environmental footprint.
It is worth remembering that from lithium to solar, all technologies were expensive at first: When solar energy was first launched it was very expensive but the global price of solar has plummeted by 89% from $359 to $40 per megawatt hour from 2009 to 2019 (Toussaint, 2024). History teaches us that technology becomes cleaner, better, and cheaper as both internal and external economies of scale are exploited. Thus, I would argue that in the long term, the potential reduction in emissions in other sectors, driven by AI, can offset the environmental impact caused by the short term issue of use and training these models. AI also offers significant potential for decarbonization, efficiency, and sustainability across multiple industries. In the secondary sector, supply chains can be optimised via Just-In-Time inventory management. This includes demand forecasting, where patterns in historical data, trends, and external factors can be analysed accurately. Moreover, assistance in transportation management can be put in place, as AI can be used in route optimisation. For example, following a test conducted by Google where they used AI-generated routes in test flights with the goal of reducing contrails, it was concluded that aviation emissions can be reduced by up to 20% (Economic Impact, 2024). Of course, we must be careful that we do not see AI as our saviour but instead see it as one way of complementing existing approaches and technologies - the hope is that alongside these, AI will help us achieve our Net Zero goals and limit heating to a maximum of 1.5 degrees as agreed at the COP21 in Paris.
Given the profound potential of AI to successfully address socio-economic challenges it is vital to consider its development and regulation to ensure AI remains a blessing over the long term. One major concern is AI’s impact on individuals, as millions of middle-income jobs, especially in manufacturing and creative industries are at risk of being displaced by generative AI and automation. In fact, in the USA alone, about two-thirds of occupations may be subjected to AI automation (Goldman Sachs, 2023) and South Korea has already replaced over 10% of its workforce with robotics (HR Asia, 2024). However, government policies to retrain and upskill workers can offset some of the fallout from this. Examples of this can already be seen as Singapore began a subsidy program in order to aid those over 40 to gain skills and knowledge in AI models (Unnikrishna, 2024).
Whilst concerns about generalised AI may or may not be farfetched, the immediate challenge around AI is to ensure that current and future users of it really understand how to use it ethically and efficiently, being wary of its power to hallucinate, to manipulate and to mislead. The solution to this will be a mixture of government legislation (e.g. the EU AI Act banning social scoring, the US AI Act banning deep fake pornography) to sector self-driven initiatives (e.g. Facebook making it clear if a political ad is AI generated) to education in schools (e.g. Australia’s AI in Education policy). This goes to show that addressing AI’s impact on society is not just a challenge of economic transition but also one of reeducating our world’s populations. As Alvin Toffler once said: "The illiterate of the 21st century will not be those who cannot read and write, but those who cannot learn, unlearn, and relearn.”
Bibliography:
6D Model of National Culture
AI-driven crop yield prediction and disease detection in agroecosystems
AI poised to drive 160% increase in power demand
Artificial intelligence breakthroughs in pioneering early diagnosis and precision treatment of breast cancer: A multimethod study
Artificial intelligence is helping improve climate models
Credit Suisse Say on Average Adults
EU Parliament Report on AI
Generative AI could raise global GDP by 7 percent
Google AI emissions
The price of solar electricity has dropped 89% in 10 years
Singapore’s Bold AI Policy Offers Subsidized Education for Adults 40+
HR Asia
Impact of AI on Emissions