From colleagues to Cobots: How workplace AI rewrites culture for better and worse

This study synthesizes empirical evidence on how workplace artificial intelligence (AI)—ranging from collaborative robots to algorithmic management—restructures organizational culture, identity, and power dynamics. A PRISMA-guided systematic review and meta-analysis of seven databases and grey literature identified 88 empirical studies conducted between 2015 and 2025. The evidence was integrated through thematic synthesis complemented by a descriptive quantitative summary. The findings converge on five key themes: AI simultaneously threatens and enables work identities; it reconfigures social dynamics by fostering collaboration while also introducing strain; it shifts power toward data-driven control and surveillance; it accelerates human–AI augmentation and job redesign; and it produces uneven cultural adaptation characterized by fear, resistance, and fragile trust. Outcomes are consistently moderated by factors such as leadership quality, organizational support, participatory implementation, and system transparency. The evidence indicates that AI is not culturally neutral; its cultural consequences depend heavily on governance and implementation choices. Practically, organizations should implement transparent communication strategies, involve workers in co-design processes, establish robust ethical and privacy safeguards, and promote targeted upskilling and AI literacy. These measures aim to translate efficiency gains into cultural value while safeguarding autonomy and well-being. Such approaches foster trust and support sustainable performance and cultural resilience in AI-enabled workplaces.