Meet WELL-AI: A Groundbreaking Conversational AI for Mental Wellbeing
WELL-AI makes mental wellbeing support equitable, adaptive, and accessible to all, in our first case study especially young people facing modern stressors like climate anxiety, academic pressure, and job insecurity.
Unlike conventional wellness apps, WELL-AI adapts its support to each user’s unique capabilities: what they value, what they can do, and the real-life constraints they face. To that end, we rely on two powerful theories: the Capability Approach and Action Control Theory.
WELL-AI is not another chatbot. It is a conversational companion app, available online and as a mobile application, powered by the most advanced AI techniques, crowdsourced citizen wisdom, and proven scientific frameworks. It personalizes its guidance based on whether a user is more likely to act quickly on advice or needs step-by-step planning to bridge the gap between intention and action. This ensures that every piece of support it offers is relevant and actionable, tailored to the individual’s life circumstances, goals, and digital literacy.
By turning real-world wellbeing strategies into a smart, accessible dialogue, WELL-AI empowers users to build psychological resilience before crises arise. Its long-term vision is even broader: to provide open-source tools, inform national mental health policies, and establish Finland as a global leader in ethical, inclusive AI for mental wellbeing.
We are building the future of mental wellbeing: equitable, evidence-based, and human-centered.
Early concept UI sketches
NB: Nothing final here. Just ideating.
Latest Results
Alorwu A, Van Berkel N, Moilanen J, Sharmila P, Meftahi N, Yatani K, Visuri A & Hosio S (2026)
Crowd-Powered Discovery of Mental Health Self-Care Techniques in Higher Education
Transactions on Computer-Human Interaction (TOCHI), to appear
Ali M, van Berkel N, Tag B, Paananen V, Oppenlaender J, Yatani K, Hosio S (2025)
Investigating Mental Wellbeing Self-Care in Higher Education Using BERTopic Modeling
Discover Mental Health, accepted for publication