How Soulr uses AI

Your Soul Companion is a private, calm guide that helps you reflect on who you are, what you value, and what you're looking for. It listens, remembers what matters, and offers perspective when invited.

The Companion uses leading large language models, accessed through Soulr's secure gateway. Your conversations are never used to train public AI models.

When we suggest a connection, the app can show a short, human-readable explanation of why this person might be meaningful to you — grounded in values and reflections you've actually shared. You see the reasoning; you stay in control of who you reach out to.

Inside a match conversation, the Companion may offer a few thoughtful openers or follow-ups. Suggestions are optional — tap to use, edit before sending, or ignore entirely. Nothing is sent on your behalf.

As you reflect with the Companion, it can save durable "moments" — dreams, fears, goals, values, turning points — so future conversations feel like talking to someone who actually knows you.

• Memories are stored privately to your account.
• Memories are never shared with other members without your consent.
• You can view, edit or forget any memory from your Memory timeline.

The Soul Companion offers perspective. It is not a substitute for professional support:

• It is not a therapist or mental health professional.
• It is not medical advice.
• It is not legal or financial advice.
• It can make mistakes, misremember, or be wrong.

If you're in crisis, please contact local emergency services or a qualified professional.

• Turn off Companion memory at any time in settings.
• Edit or delete any individual memory.
• Clear your entire Companion history in one tap.
• Export your reflections before deleting them.

Open the Memory timeline, find the moment you want to forget, and tap Forget. The memory is removed from the Companion's context immediately and permanently deleted from our database shortly after.

AI responses are filtered for harmful content. We review patterns in aggregate (never individual conversations) to improve quality and safety. Sensitive prompts that suggest self-harm or risk to others surface supportive resources rather than open-ended advice.