An Analytical Study on the Effect of Large Language Model on the Unemployment of Software Engineers
Abstract
This dissertation examines the impact of Large Language Models (LLMs) on software
engineers' joblessness, focusing on the debate over whether LLMs lead to technological
unemployment or complement human work. The positivist, quantitative developmental study,
with a cross-sectional design, was conducted using purposive sampling to obtain software and
IT professionals and students in the amount of 63 people, who were then asked to complete a
structured questionnaire followed by data analysis using reliability test, one-way ANOVA,
Spearman correlation and multiple regression in SPSS. Research reveals that LLM use is
widespread and common, and that there is a greater ‘risk of losing job' if using LLM than in
using it if they are a student or senior employee, and that it is more common among engineers
with less than one year of experience. The number of times they had used LLM was not the
best independent predictor of risk perception, as was familiarity with the tools rather than
experience. The results confirm a task substitution rather than whole occupation displacement
trend and suggest that early career engineers, employers, and educators be targeted for LLM
literacy to cope with the risk of employment due to continued industry-wide uptake of LLM
use.
Collections
- 2026-2030 [51]