By Aidan West
Automation of tasks is always a disruptive process but is a beneficial consequence to rising productivity. As research in both history and economics show, there is no intrinsic conflict among technological change, full employment, and rising earnings. Through innovation, the efficiency and veracity of work that is able to be accomplished increases greatly. In turn, a virtuous cycle is achieved: increase in innovation leads to an increase in productivity which provides society with the necessary resources to invest in the disenfranchised whose livelihoods are disrupted by the changing structure of work.
But AI is a unique technology due to its complexity. AI combines large amounts of data with supernal processing power that allows its software to automatically learn any pattern or feature of that data. It is a broad field of study including the following major subfields: machine learning, deep learning, computer vision, and natural language processing. Eventually, AI could replicate many human traits through this type of learning. However, at its current state of evolution, it is only capable of replacing low-skill labor jobs and performing highly specialized tasks. Even so, the result of our current workforce is significant.
Currently, there is an observable bifurcation of work that is in part due to AI. The growth of AI has led to the demand of hybrid jobs which combine both technical skills and soft skills. Taking this concept further, superjobs require the combination of technical skills and soft skills while also combining the skills of other traditional jobs. This type of work is being observed as having the fastest acceleration in wages but requires multilateral redesign of the workplace to introduce AI to these traditional roles. Ultimately, these jobs are only available to the highly educated populace constituting the upper class while the middle class continues to decline and the lower class suffers.
An observable trend is the automation of services such as self-service checkouts once occupied by workers. More evidence is also being shown for automation and self-service in a variety of other sectors. Accounting, law, financial planning, lending, investment management, medical diagnostics and call centre service are all becoming more and more artificial. It is likely that vehicles in the near future will be able to drive themselves due to a new wave of AI technology. Effectively, any job that is low-skill and able to be replaced by AI will be.
Yet the trend of low-skilled laborers being replaced by AI isn’t as prevalent as it ought to be based on the rapid advancement of AI. This is because the speed of the transition to complete automation of standard jobs is only partially dependent on technological progress. Companies will generally attempt to save as much money as possible and replacing workers with advanced AI is not the most cost-efficient. The automation currently provided in many places consists of technology that mimics the role of workers without necessarily making the interaction more efficient. This slightly reduces the costs spent by companies but doesn’t increase their productivity by a significant margin. The usual displacement effect occurs with workers replaced by this technology without the benefit provided to other workers due to an increase in productivity.
AI constitutes a significant part of our modern economy and naturally the field will continue to grow to its natural precipice. It is an extremely useful but frightening invention due far-reaching capabilities. It learns instantaneously, but only through the algorithms it’s filtered through and the data it is provided. Just as programmers must account for human bias in AI development, politicians must also account for the effects of this technology on citizens’ lives.
