Advances in technology are almost always considered positive because they increase productivity and ease frustration of completing simple, menial tasks. However, technological improvement does not come without cost; some tech advancements can result in the automation of jobs that used to be performed by humans.
Currently, automation is occurring more in service and manufacturing positions. It is one factor that contributes to labor market polarization, or the disappearance of middle-income, routine task type jobs in the U.S. Despite a strong labor market, fear of job loss from automation is common. Some parents with young children already worry that robots will take all potential jobs for their children in the near future.
The short-term impact of automation is not directly observable, but two economists at the University of Oxford, Carl Benedikt Frey and Michael A. Osborne, attempted to quantify jobs that are at risk in a 2013 article about automation. The researchers used a machine-learning algorithm to estimate the probability that an occupation will become automated in the next few decades. The probability is calculated based on three major factors of the occupation: perception and manipulation, creative intelligence, and social intelligence. The result can be interpreted as the likelihood that engineers will be able to produce machinery that performs tasks required in each occupation.
The economists found that 47 percent of jobs in the U.S. are at risk of becoming automated. Jobs that are repetitive or routine-intensive have the highest probability of this happening. The economists predicted that the transportation and logistics occupations, as well as the office and administrative support occupations, will lose the most jobs to automation over the next decade or so.
The Oxford economists predicted that automation will occur in waves, first replacing routine tasks, then slowing as engineers reach a technological plateau. A good example, provided by Frey and Osborne, is that paralegals and legal assistants (which are considered relatively low-skill, routine-based occupations) are seeing their jobs quickly becoming automated; however, it will be a long time before computers are advanced enough to replace lawyers (whose jobs are considered high-skill, nonroutine).