How Habit keeps your data and personal information safe

Honest self-reflection only works if you trust the space you're reflecting in. This article explains how Habit keeps your personal information and journal entries private while still using AI to generate insights.

The short version

• Your name, email, account details, and anything else that could identify you stay on Habit's servers.
• Only your check-in answers go to the AI provider.
• Those answers are stripped of identifiers first. Then they're batched and interleaved with other users' answers. The AI provider can't tell which insight came from whom.
• Habit never uses generated data to train AI models.
• You can delete individual entries or your entire journal history at any time.

What stays on Habit's servers

When you create an account, we collect the information needed to run the service: your email address, account settings, subscription status, and payment history if you upgrade. We also store the outputs of your check-ins: your saved insights, patterns, reflections, and small steps.

All of this is tied to your account and stays in Habit's secure infrastructure. It isn't shared with the AI provider or any other third party for processing.

What the AI provider actually sees

When you complete a check-in, the AI generates an insight based on your answers. The AI provider receives only what it needs to produce that insight:

• The answers you selected during the check-in.
• Context needed to shape the response, such as the type of insight being requested. None of this context is tied to your personally identifying information. No location, date, time, or account details are sent to the LLM.

The AI provider does not receive:

• Your name or email address.
• Your user ID.
• Your account history or subscription status.

How your insight is created

1. You answer the check-in questions on your device.
2. Habit's server receives your answers and stores them with your account.
3. When an insight is needed, Habit pulls those answers, removes anything that could identify you, and sends the anonymous response to an AI provider.
4. The AI provider generates text based only on the anonymous answers it received.
5. Habit matches the generated insight back to your check-in and shows it to you.

From the AI provider's perspective, every request looks the same: anonymous check-in answers with no names, dates, locations, or account information attached. The matching back to your account happens only on Habit's servers.

How Habit improves its AI

Habit continually researches ways to make its guided check-ins and insights more useful. Our improvement work happens on our own systems. We use synthetic test data and carefully anonymized interaction patterns, not the content of your personal reflections.

What we don't use for research

• The text of your insights or reflections.
• Your individual check-in answers.
• Any data that could be linked back to your account.

What we may use for research

• Aggregated, anonymized signals, such as which question paths users complete most often.
• Synthetic examples created by Habit to test prompt quality and insight formats.
• Feedback you explicitly choose to share with us.

What we can't control

We send only anonymous, interleaved check-in answers to the AI providers that power Habit's insight generation. We don't give them permission to train on that data, and our agreements prohibit it. But we can't fully audit or guarantee what happens inside a third-party provider's infrastructure. If you want to avoid any possibility of an external LLM processing your reflections, don't use AI-generated insights.

Your raw reflections always stay on Habit's servers under your control.

Encryption and storage

Data is encrypted in transit using HTTPS/TLS. Stored data is protected by the security practices of our hosting infrastructure, including access controls and encrypted storage volumes.

Only a small number of authorized Habit team members can access production systems, and only for operational support. No one at Habit reads your individual journal entries as part of product development, marketing, or research.

What you can control

• Delete an entry: remove any individual check-in from your journal library.
• Delete your journal history: wipe all your entries from your account settings.
• Delete your account: request full account deletion, which removes your account data and journal history from our servers.

Deleted data is permanently removed. We don't keep hidden copies.

What this means for you

You can use Habit knowing your reflections aren't being mined, sold, or fed into public AI training. The AI helps you interpret your answers without knowing who you are.

If you want to see exactly what we collect and how we handle it, read our privacy policy.