Discernment
The capability behind every response our AI gives – in every setting below.
Kwalia is a research company building AI that helps people deepen their understanding of the situations they’re part of.
Our approach has its roots in clinical practice and philosophy – and has been refined through sustained work with AI.
Read the thesis ↗Our co-founder Paul spent his entire career developing IRH (Integrative Relational Health) – a framework that draws on clinical practice and post-Cartesian philosophy to explain how understanding develops between people.
Since starting Kwalia, we’ve examined how AI responds to hundreds of real situations – and where it hands people conclusions without helping them reach their own.
Those conversations helped us identify where AI cuts understanding short – and develop principles that produce a different kind of orientation.
Knowing what to say, how and when – to give understanding the best chance to develop.
Discernment starts in a single conversation. Our ambition extends to wherever people work, learn and decide together – and to how AI’s progress itself is measured.
The capability behind every response our AI gives – in every setting below.
Helping one person understand a situation they’re part of – a decision, a relationship, a piece of work.
Helping people understand together – in teams, classrooms and organisations, where no one sees the whole picture alone.
Research grounded in individual and collective settings, so human understanding becomes a measure of AI progress.

3x founder, engineer and product leader with 15 years in tech. Built and scaled products across social, mobile gaming and AI – including two chart-topping games, a social platform that grew to 10M+ users and a funded AI-native shopping platform. As a startup advisor, he’s helped founders find product-market fit and raise over $15M.

Clinical psychologist with 35 years across therapy, supervision and teaching. Founder of the Center for Integrative Relational Health (IRH) and architect of the IRH framework for how understanding develops between people. Author of “Ways of Knowing” and host of “Etymologies of Care”.
We’d like to hear from researchers and builders working on model behaviour, evaluation, and what AI must become next.