Abstract
Artificial intelligence (AI)-driven healthcare is reshaping clinical work, not only through “adoption” of new tools but by reconfiguring how work is organised, altering workflows, clinical judgement and accountability. Yet clinicians’ responses to AI are not uniform, with similar AI-enabled environments producing divergent experiences of workload, trust, risk and willingness to use AI (Choudhury et al., 2022; Shamszare et al., 2023). These divergences matter for workforce sustainability because they can influence doctors’ wellbeing, job satisfaction, engagement and career intentions.
This developmental paper theorises career impacts of AI-driven healthcare using Conservation of Resources (COR) theory, which posits that individuals strive to protect and build valued resources and experience strain when resources are threatened, lost or when investments fail to yield gains (Hobfoll, 1989; Halbesleben et al., 2014). We propose that AI integration is experienced by doctors as shifts in valued personal and work resources (e.g., time and cognitive bandwidth, autonomy and discretion, confidence in judgement, collegial support, and clarity of accountability). Where AI increases accountability burdens, resource investment or uncertainty about clinical authority, it may be appraised as resource threat; where it improves decision support, reduces friction, or enables learning, it may be appraised as resource gain. These appraisals are expected to initiate loss or gain spirals that shape career outcomes over time (Halbesleben et al., 2014).
Professional identity is positioned as a key personal resource mechanism within COR: it supports meaning, legitimacy and expert judgement (Caza & Creary, 2016). AI may threaten identity when it is perceived to undermine epistemic authority or autonomy and strengthen identity when experienced as capability-augmenting. Ethical and epistemological debates about black-box systems highlight the centrality of transparency and trust in medical AI (Durán & Jongsma, 2021), and emerging empirical work suggests AI can reshape professional autonomy and identity in medical specialties (Lombi et al., 2024). We further advance a career-stage perspective, arguing that early-, mid- and late-career doctors differ in resource reservoirs, perceived return on reskilling investments and vulnerability to loss spirals. Finally, we conceptualise organisational learning culture as a resource “passageway” that conditions whether AI implementation becomes capability-building or disruptive by enabling training, feedback, support and shared sensemaking (Hobfoll, 2012). We outline an exploratory sequential mixed-methods programme (interviews followed by a two-wave survey) to refine constructs, finalise validated measures and test the proposed moderated relationships, generating actionable organisational levers for sustainable AI-enabled clinical work and doctors career sustainability.