The robot revolution has long been seen as an eschatology for manual workers who perform passive and repetitive tasks. McKinsey’s research, widely cited in 2017, shows that 50% of work activities can already be automated using the latest technology, and these activities are the most widespread in the manufacturing sector. New data shows that white-collar workers, even those who assume more analytical thinking, higher wages and relative job security, may not be sure due to constant automation.
A new study by economist Michael Webb of Stanford University found that artificial intelligence was discovered (powerful computer technology, such as machine learning, that can make similar decisions to humans and learn and improve using real-time data ). Published by the Brookings Institution. The range of jobs potentially affected by AI goes beyond white-collar jobs, such as telemarketing, a field that has already been extinguished by bots, professions, chemical engineers, physicists and markets that were previously considered square in the human sphere . Survey Analyst
The new study examines pairs of overlapping nouns and job descriptions in AI patents to identify jobs most affected by AI technology. For example, job descriptions of relatively common market research analysts with high rates of AI surveys share many terms in common with existing patents. Similarly, the markets for “data analysis”, “marketing tracking” and “identification”.
More prospective than other studies in that it analyzes the patents of technologies that are not yet fully developed or distributed.
In general, studies have shown that automation effectiveness estimates for significantly different personnel have focused on tasks that can be automated using existing techniques. For workers with low wages and low education, where robotics and software often eliminate part or all of a specific occupation, this result is usually the most damaging.
The ghost of automation has raised concerns about how many Americans can support themselves when jobs are mechanized and whether the loss of low-income jobs will increase wealth inequality. According to this new patent study, the impact of automation is much broader and can affect the highest paid office workers.
Warning: Some AI patents may never be used and cannot be used with the initial intention. In addition, the actual work is not completely defined by the text of the description of the original work. However, this study provides a framework for viewing common exposures to automation.
Adam Ozimek, chief economist of the Freelancing Upwork platform, says: “For example, patenting a device that conducted market research using artificial intelligence does not mean that AI will really succeed for real commercial use. ”
The Stanford study also says that these workers don’t really know if they will lose their jobs, but they can only affect their work. Therefore, it can be used to increase work on behalf of these techniques.
White collar jobs are not the first to be at risk in technology. Many of the white collar jobs that must be sent abroad, such as actuaries, technical writers, customer service representatives, have not seen the cuts expected 10 years ago.
And for what it’s worth, market research analysts seem to think they can’t do many elements of their work using artificial intelligence.
Gina Woodall, president of the market research firm Rockbridge Associates, told Recode: “I know what is happening, but I can’t understand why.
Opinion of the market research analyst
According to the Bureau of Labor Statistics, about 700,000 in the United States say that market research analysts, people studying market conditions to sell products or services, and job growth prospects in the next decade are much more High than average. As the title suggests, it generally leads to analytical thinking about disparate data, so it is generally safe to talk about automation.