How does an AI generate insights from my moods?
By Siri Bandagiriye · It looks for patterns over time, such as which days you seem lower or what comes up alongside better days. Treat any insight as a question to check against your own experience, not a verdict.
It looks for patterns across time: what you talk about, when you seem lower, what comes up alongside better days. A pattern might be "Sunday evenings are hard" or "days with a walk feel lighter".
The value is in noticing, not in the AI's conclusions. Treat any insight as a question to check against your own experience.
If it doesn't fit, it's wrong, and you can ignore it.
What the research says
In a 2020 study, machine-learning models using passively collected smartphone data predicted students' depressed mood hour to hour, though the sample was only 31 people.1
Written by Siri Bandagiriye
I've been a Buddhist monk and mindfulness teacher for 30 years, and for six years a chaplain with Correctional Service Canada, sitting with people in crisis. I built Saati in Halifax as a companion for the ordinary hard days. If you'd like to see what it's like, it's free to try.