Artificial Intelligence Results in Layoffs in San Francisco

By now, most people have a basic understanding of what artificial intelligence is and how it can be used in our daily lives; however, what most people are unaware of is how it can adversely affect the way we work. Recently, California began tracking how artificial intelligence can affect the job market in one of the biggest job markets in the world: San Francisco.

Although this study was a small one concentrated in a region saturated with employees working in the technology industry, it uncovers something interesting: a college degree will not prevent you from being replaced with artificial intelligence. This may come as a surprise to most people who are told, from a very young age, to study hard and get a college degree so that they can have job security. The idea of college degrees providing young people with job security has been turned on its head.

The study, which was spearheaded by the California Policy Lab at the University of California and the California Employment Development Department shows that college-educated workers in tech-exposed fields had filed significantly more unemployment claims after the ChatGPT-3.5 launch in 2022. Examples of jobs in tech-exposed fields include: administrative support occupations, paralegals, legal assistants, and computer programmers. In fact, unemployment claims from this group continue to remain high through May of 2026, for a total of four years now. The statistics are clear: from November 2022 and July 2023, unemployment claims from workers with a bachelor’s degree rose from 13,000 to 22,000 claims per month. It nearly doubled.

One factor that may contribute to this is the fact that these sectors have business models that can incorporate artificial intelligence. Clearly, it would be easier for an office to incorporate artificial intelligence than a plumbing business. Some jobs just require a person. Thus, employers in tech-exposed fields are more likely to utilize artificial intelligence in their business. In addition, employers in these sectors can also integrate artificial intelligence more seamlessly into their business, meaning the artificial intelligence will permanently replace a worker. In fact, in another report issued by the McKinsey Global Institute, the Bureau of Labor Statistics stated that “[s]oftware innovations have automated many of the tasks performed by bookkeeping, accounting, and auditing clerks...As a result, the same amount of work can be done with fewer employees, which is expected to lead to job losses over the projections decade.”

A limitation in this study is that this study takes its information from the unemployment claims filed. The individual workers who are college-educated may very well find another job, completely exit the labor force, or become self-employed. If this happens, they would not be reflected in this study because they would not have filed for unemployment benefits.

Clearly, artificial intelligence will have some impact on how employers run their businesses and on the types of jobs that can be automated. Only time will tell how the employer and employee dynamic changes with the rise of artificial intelligence. Employers have a huge incentive to limit the number of employees required to run a business, while employees have a huge incentive to preserve their jobs. Artificial intelligence adds a new factor to the mix: what employees can be replaced by AI?