On March 21, OpenAI, the company that invented the recently popular large-scale model chatGPT / GPT4, and Binzhou University and other institutions published a preprint of a paper, aiming to investigate the potential impact of large-scale models such as GPT on the job market.
Here is the abstract of the paper:
We investigate the potential impact of generative pretrained transformer (GPT) models and related techniques on the U.S. labor market. Using a new criterion, we assess how well occupational competencies correspond to GPT, while incorporating human expertise and GPT-4 classifications. Our research found that approximately 80% of the U.S. labor market could be impacted by the introduction of GPT with at least 10% of their work tasks affected, while approximately 19% of workers are likely to see at least 50% of their work tasks affected . The impact spans all pay levels, with higher-paying jobs likely to be more at risk. It is worth noting that this effect is not limited to industries with higher recent productivity growth. We conclude that generatively pretrained transformers exhibit the characteristics of a general technique (GPT), suggesting that these models may have important economic, social, and policy implications.
Paper Link:
Paper