SaatiLearn

Safety and privacy

Is it safe to rely on AI for mental health support?

It's safe to lean on, not to rely on alone. An AI companion can support everyday stress and lonely evenings, but it shouldn't be your only support, especially if things are serious or getting worse.

Safe to lean on, not safe to rely on alone. An AI companion can be a good support for everyday stress and lonely evenings. It should not be your only support, especially if things are serious or getting worse.

A safe app is honest about its limits, points you to crisis lines, doesn't claim to treat or diagnose, and encourages your human connections.

If an app makes you feel you need it more and people less, that's a warning sign.

What the research says

The American Psychological Association's first recommendation is not to rely on AI chatbots and wellness apps to deliver psychotherapy or psychological treatment.1

A 2025 study of 9,000 chatbot responses found the chatbots were inconsistent on suicide-related questions of intermediate risk, so they shouldn't be your only plan in a hard moment.2

Siri Bandagiriye

Written by

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.

References

  1. American Psychological Association, Health Advisory on generative AI chatbots and wellness apps (2025)
  2. Psychiatric Services, McBain et al., Alignment between large language models and expert clinicians in suicide risk assessment (2025)