D3.295 - Comparing Real-World and AI-Enabled Market Research in Caregivers of Individuals with Peanut Allergy

Understanding caregiver perspectives in peanut allergy management is vital for developing patient-centred therapies and support. Traditional market research methods are effective but can be resource intensive. Recent advances in generative AI (Gen AI) offer innovative opportunities to simulate and explore stakeholder insights using fictive AI personas.

Two parallel market research studies were conducted: one in a real-world cohort of caregivers to individuals with peanut allergy (N=20), and one using AI-generated personas (N=30), developed and interviewed via advanced natural language model (GPT-4.5). Both groups responded to a standardised interview guide focused on management approaches, daily impact, coping strategies, and support systems. Insights, diversity, and administrative workload were compared across methods.

AI-enabled research demonstrated high concordance with real-world findings; the overall narrative consistency between AI-generated personas and actual caregivers was strong, with minor variations on nuanced topics. AI-based market research delivered significant advantages in speed (~70-90% time saving), cost (up to 80% lower), and administrative burden, with additional benefits in reaching underrepresented perspectives. Profiled personas captured complex, real-life challenges—such as financial stress, stigma, and coping mechanisms—comparable to human interviews. Methodological rigor in AI-generated personas creation was essential to ensure realistic and consistent insights.

Gen AI- powered market research provides a reasonably consistent, scalable, and resource-efficient complement to traditional approaches with the potential to enable a deeper and more diverse understanding of caregiver experiences in peanut allergy management. Integration of AI and real-world methods may enhance future research and product development processes.