Abstract
In recent years, generative AI has undergone major advancements, demonstrating significant promise in augmenting human productivity. Notably, large language models (LLM), with ChatGPT-4 as an example, have drawn considerable attention. Numerous articles have examined the impact of LLM-based tools on human productivity in lab settings and designed tasks or in observational studies. Despite recent a