Macroeconomic impacts of AI
created, resulting in a net reduction of 14 million jobs globally. 5 This represents a significant structural churn of 152 million jobs, or 23% of the total workforce studied.
Beyond employment figures, AI could change the structure of the labour market itself. As AI becomes capable of performing tasks previously performed by office workers, the value of traditional cognitive labour is diminished. The assumption that ‘human intelligence is scarce and expensive’ is now being challenged. A single language model can perform tasks like writing legal summaries or debugging code at a near-zero marginal cost. 6 This means fewer, already experienced humans are needed to oversee AI models, which changes how people are recruited and developed over time. When speaking with the CEO of MacroHive, an AI-backed financial research firm, he noted that while financial analysts will always exist in banks, ‘their involvement will be scaled down’. This is of particular concern to economies like the United Kingdom, where 87% of jobs are in the service industry and companies like MacroHive are making traditional human analysis obsolete. In previous technological revolutions, displaced workers transitioned into newly created jobs, but the new roles created by AI, such as data science or advanced programming, demand highly specialized technical skills that are not suitable for everyone. Without the infrastructure to retrain this large segment of the population, they could face prolonged structural unemployment or underemployment. This would mean lower wages for approximately 40% of jobs globally and a displacement of between 3-14% of workers in developed western economies. 7 This widening skill gap has serious implications for economic equality. The benefits of AI are likely to fall upon a select few who own or control AI infrastructure, alongside a highly skilled group capable of using AI tools to enhance their productivity. A summit of top AI executives, including DeepMind's Demis Hassabis and OpenAI's Sam Altman, concluded that without significant intervention, AI is likely to ‘exacerbate increasing wealth and income inequality within countries’ and may even lead to the ‘political dominance of wealthy individuals and corporations’. 8 This perspective suggests that the existing social construct, where people receive a stake in society in return for their labour, is in jeopardy. Part II: the mitigating benefits and economic catalysts of AI Despite significant challenges, the transformative benefits of AI have the potential to outweigh its short- term negative consequences, beginning an era of unprecedented economic growth and improved quality of life. This perspective is shared by some of the most prominent leaders in the AI space. Jensen Huang, 5 Future of Jobs Report 2023 | World Economic Forum. https://www3.weforum.org/docs/WEF_Future_of_Jobs_2023.pdf 6 AI Is Cheap Cognitive Labor and That Breaks Classical Economics – online discussion forum at https://www.reddit.com/r/artificial/comments/1kq9pvb/ai_is_cheap_cognitive_labor_and_that_breaks/. 7 International Monetary Fund: Will AI Transform the Global Economy https://www.imf.org/en/Blogs/Articles/2024/01/14/ai-will-transform-the-global-economy-lets-make-sure-it- benefits- humanity#:~:text=Roughly%20half%20the%20exposed%20jobs,of%20these%20jobs%20may%20disappear. 8 The AI Summit Where Everyone Agreed on Bad News. https://time.com/7313344/openai-google-deepmind-summit-social-contract-inequality/.
86
Made with FlippingBook - PDF hosting