Artificial Intelligence Adoption and Employee Productivity in Nigerian Organizations: Evidence from Seplat Energy, Eket

Abstract

This study investigates the relationship between Artificial Intelligence (AI) adoption and employee productivity within Seplat Energy, an indigenous Nigerian energy company, focusing on its Eket operational base. Despite rising AI investment across Nigeria's oil and gas sector, independent empirical evidence on whether such investment translates into measurable employee productivity gains remains scarce. Drawing on the Technology Acceptance Model, Task-Technology Fit theory, and Diffusion of Innovation theory, the study examined three dimensions of AI adoption, automation of routine tasks, AI-assisted decision support, and employee perception/readiness for AI as predictors of employee productivity. A descriptive survey design was adopted; from a population of 190 staff, a sample of 128 was drawn using Taro Yamane's formula and stratified random sampling. A validated structured questionnaire (Cronbach's alpha 0.79-0.87) was administered, yielding 120 usable responses (93.8% response rate). Data were analyzed using descriptive statistics and the Chi-square test of independence at p < 0.05. Automation of routine tasks (χ² = 10.594, p = 0.001) and AI assisted decision support (χ² = 5.672, p = 0.017) were significantly associated with employee productivity, while employee perception/readiness was not (χ² = 1.558, p = 0.212). The findings suggest that task-embedded dimensions of AI adoption exert a more direct influence on productivity than general employee attitudes, with implications for how Nigerian energy companies sequence training and change-management investment. Note: given constraints on direct access to live organizational survey deployment, the dataset analyzed is a realistic, internally consistent illustrative dataset constructed to demonstrate the full analytical procedure; it should be replaced with field data prior to any policy application.