Researchers at the University of Lausanne have addressed two questions that are often asked in the context of our overall societal gainful employment mantra when it comes to the future of intelligent robots:
- How many jobs will be eliminated by further automation?
- How can I make sure that my workplace is not one of them?
The research design
To identify the jobs that could be replaced most by robots, the team first created a list of robot capabilities taken from Europe’s multi-year H2020 Robotics Roadmap, a collaborative effort between the European Union and the robotics industry. They then sifted through research papers, patents, and descriptions of commercially available robots to determine how mature each robot capability was.
These were then matched with human skills described in the Occupational Information Network (O*NET) dataset, which contains details on nearly 1,000 occupational profiles. By assessing how many of the skills required for a particular job can be taken over by a robot or could be taken over in the distant future, the team was able to determine which occupations are most at risk of automation.
The context
Whereas earlier waves of automation primarily affected low-skilled jobs, the rapidly improving capabilities of machines mean that middle- and high-skilled occupations are also increasingly at risk.
The pace of progress also means that jobs can change much faster than they used to, resulting in workers* needing to retrain and acquire new skills several times over the course of their lives. We had cited this in many previous piqd’s.
The result
It ranked the approximately 1,000 O*NET occupations on how susceptible their job was to automated robot work. The values for one’s own profession can be run through once on a specially set up website.
In contrast to the Job Futoromat from the IAB, for example, this self-test allows you to see for yourself which possible job changes would be conceivable for employees at risk – based on the previous skillsets that were assigned to the old occupations. These occupation suggestions are generated with a proprietary algorithm.
By calculating the similarity of requirements in two occupations, researchers were able to find a measure of how much effort it would take for workers* to retrain. They then combined this with the automation risk of each job to determine the job to which a worker could most easily transfer without running the risk that the new job would also soon become redundant.
My classification
Such mechanistic research is certainly helpful as a thought experiment, but it itself follows an algorithmic mindset that can hardly do justice to the multi-layered complexity of an increasingly digitally literate society.
Viewed positively, such a tool can possibly help to increase personal and social resilience, because one is directly shown professional alternatives.
But I don’t know if it helps to lift the necessary self-reflection, to listen into oneself and to ignite a motivation to learn that is also conducive to personal development. Quite apart from the gainful employment mantra, which, for social security reasons, does not want to accept when people precisely do not want to live in the export-oriented white-collar mode and devote themselves to the common good at the most diverse levels.
Article published on piqd on July 30, 2022 as a reference to the SingularityHub article There’s Now an Algorithm to Help Workers Avoid Losing Their Jobs to an Algorithm
