According to the study, GPTs have the potential to significantly affect a diverse range of occupations within the US economy, demonstrating a key attribute of general-purpose technologies.

Also Read: GPT-4 devises plan to ‘escape’ by gaining control of a user’s computer

The study identifies the potential exposure that each job has to large language models. In simple terms, exposure refers to whether using a GPT or GPT-powered system can reduce the time it takes humans to do their jobs by at least 50 percent. The higher the exposure a job has, the higher its chances of getting automated.

Jobs at risk of automation due to GPT

According to the study, higher-paying jobs are more likely to be automated compared to lower-paying ones.

The study also discovered that professions that are heavily reliant on scientific and critical thinking skills are less prone to automation. Conversely, jobs that require proficiency in programming and writing are more susceptible to being automated.

The authors of the study identified specific jobs that are at high risk of automation such as Web and Digital Interface Designers, Blockchain Engineers, Tax Preparers, Writers and Authors etc.

Here’s a list of all the jobs that could be automated due to Large Language Models.

Interestingly, the authors also revealed the names of 34 jobs that have significantly lower exposure to GPT-powered models, meaning there is a very faint chance of them getting affected due to LLMs.

Also Read: GPT-4 takes the world by storm – List of companies that integrated the chatbot

The list of 34 jobs includes professions such as athletes, cooks, mechanics, helpers, tapers, tire repairers, etc.

Here’s the list of jobs that are least likely to be automated:

View Full Image

This is not the first time that OpenAI has considered the possibility that its offerings could lead to the loss of many people’s livelihoods. Just last week, OpenAI CEO Sam Altman conceded that ChatGPT could eliminate a lot of jobs, but insisted that it would also create a lot of new jobs that are hard to imagine today.