Employee silence and voice in AI-driven workplace learning: an integrative literature review

Purpose This study aims to review how employee silence and employee voice shape AI-driven workplace learning. It focuses on the conditions under which employees withhold concerns, questions and learning difficulties or speak up during AI-related training, reskilling, automation and digital transformation. Design/methodology/approach An integrative literature review was conducted using a systematic OpenAlex search and screening protocol. The search combined terms on artificial intelligence, automation, digital transformation, workplace learning, training, reskilling, employee silence, employee voice and related mechanisms. After deduplication and screening, 61 studies were retained: 50 were examined in full text, while 11 were retained for limited theoretical or contextual use on the basis of abstract and bibliographic information. Based on thematic proximity, the corpus comprised 12 closely aligned studies and 49 adjacent mechanism studies. Findings This review shows that the direct literature on employee silence and voice in AI-driven workplace learning is emerging and fragmented. Most studies address the issue indirectly through adjacent constructs such as trust, job insecurity, resistance, psychological safety, participation, digital skills and AI adoption. The synthesis identifies four employee response patterns: silent compliance, defensive silence, developmental resistance and learning voice. Developmental resistance is reported as a review-derived conceptual extension because no retained study names or operationalises it directly. Originality/value This study’s primary contribution is to conceptualise employee silence and employee voice as human resource development (HRD) mechanisms within AI-driven workplace learning, rather than only as general organisational behaviour constructs. It organises the fragmented literature into four employee response patterns and integrates them in a silence–voice–learning framework linking AI-related learning conditions, employee responses, HRD interventions and learning outcomes. The framework combines the established construct of defensive silence with three review-derived categories: silent compliance, developmental resistance and learning voice. These categories extend silence and voice theory into AI-driven workplace learning and provide an agenda for empirical testing.