Generative Artificial Intelligence and Labour: Displacement and Augmentation at the Task Level

Generative artificial intelligence has extended automation from routine physical and computational activities to language-intensive cognitive tasks and other areas of human labour. From the perspective of task orientation, this paper examines changes in labour due to the above reasons and focuses on technological displacement and labour augmentation. Based on the above literature, recent theoretical and empirical studies have investigated occupational exposure, workplace productivity, employment adjustment, wage distribution and skill demand. Analysis shows that generative AI is unlikely to change all parts of a job in the same way. Standardised and codifiable tasks are more likely to be substituted; those that require contextual judgement, interpersonal communication, accountability and verification are more suitable for human labour. Evidence from professional writing and customer support indicates that AI-assisted work can increase efficiency for new employees at the company, but it is not ideal everywhere. The paper concludes that the first wave of labour-market changes will be task reallocation and employment redistribution, not universal occupational displacement. Effective adaptation therefore depends on training, institutional support, and access to complementary skills.