Aims
Artificial intelligence (AI) models using routine healthcare data are increasingly developed for dementia detection, but stakeholder acceptability remains unexplored despite being critical for implementation. We aimed to explore stakeholder perspectives on AI-driven dementia detection and its role in clinical care.
Methods
We conducted semi-structured interviews with 30 participants across three groups: healthcare workers, people with lived dementia experience, and the general public. Transcripts were thematically analysed in NVivo15.
Results
Dementia diagnosis was presented as emotionally and ethically complex, shaped by stigma, uncertainty, and unclear pathways to assessment and support. Clinicians and people with lived dementia experience saw value in earlier detection, while those without direct experience had divided views. Some valued the chance to plan and maintain control, while others saw early knowledge as distressing or futile. These tensions extended to attitudes toward AI-driven dementia detection, often mirroring broader views on dementia diagnosis. Most participants were cautiously optimistic, accepting a “powerful black box” if accurate, clinician-reviewed, and not used in isolation. AI’s perceived objectivity was seen as useful in navigating sensitive conversations around dementia risk. Across groups, participants emphasised the need for clear purpose, defined responsibility, and appropriate follow-up pathways for AI-generated dementia risk estimates.
Conclusions
Acceptability of AI-driven dementia detection using routine healthcare data depends not only on technical sophistication, but also on how such models are governed and integrated into clinical care. These may have greatest value when positioned as a clinical decision support tool within pathways that recognise the ethical and emotional realities of dementia diagnosis.