Inward

Privacy Policy

Effective: 2026-06-23

The gist

Inward is an app for self-discovery. To work, it has to handle some genuinely personal content — your reflections, your answers, your conversations with our AI guide (Sherp). We take that seriously.

Here's what's worth knowing up front:

Your reflections are yours. We don't train AI models on what you write. Our AI providers (Anthropic, OpenAI) are contractually prohibited from using API content for model training. Inward itself doesn't train models on user content either.
Sherp transcripts and AI prompts are sent to PostHog. PostHog is the analytics service we use to monitor AI quality and cost. The full content of your messages to Sherp, and Sherp's responses, are logged there. We need this to spot regressions and bugs. We don't sell any of it.
No ads. No selling your data. Inward is a subscription product. We have no incentive to monetize what you tell Sherp, and we don't.
You can delete everything. Account deletion erases your data from our database. Backups expire on a fixed schedule (see §7).

The rest of this document is the formal version.

01Who we are

Inward is operated by Inward Journeys, Inc., a Delaware corporation, with a primary contact address at 61 Richardson St, Apt PHH, Brooklyn, NY 11211. References to "Inward," "we," "us," or "our" mean Inward Journeys, Inc.

For privacy questions, contact us at privacy@getinward.com.

02Scope

This policy covers the Inward iOS app and any web property at getinward.com. It does not cover third-party services you reach through links we publish — those services have their own policies.

03What we collect

3.1 Account information

Email address (you, or Apple if you use Sign In with Apple)
Authentication tokens (managed by Supabase Auth)
Apple-relayed email address, if you choose Apple's privacy-relay option

3.2 Content you create in the app

Episode responses. Your swipe answers (Collect), open-ended answers, choices, and reactions during episodes (Learn / Reveal).
Conversations with Sherp. The full text of your messages and Sherp's replies, in both episode-bound conversations (Converse) and the ongoing advisor.
Personal File. A structured set of facts we extract from your responses to give Sherp continuity across sessions. Five categories: who you are, your world, what's happening now, your history, and how you operate.
Synthesis output. AI-generated interpretations of your responses (your "Inner Map," archetype names, growth notes).

3.3 Device and usage information

App events (which episodes you started, completed, when, on what device type)
AI events (prompt and completion content shipped to PostHog for quality monitoring — see §5)
Crash reports and errors (Sentry)
Standard server log information (IP address, user agent, timestamps) — retained briefly for security and abuse prevention

3.4 What we do not collect

We do not collect contacts, photos, location, microphone audio, camera content, or health data unless you explicitly grant a future feature access.
We do not embed third-party advertising trackers.
We do not sell or share personal information for cross-context behavioral advertising.

04How we use your information

Purpose	What we use
Run the app — generate Sherp replies, build your Inner Map, save progress	Episode responses, Sherp conversations, Personal File, account info
Monitor AI quality and cost — catch regressions, track token usage, debug latency	AI prompts + completions (via PostHog LLM Analytics)
Diagnose bugs and crashes	Crash reports (via Sentry)
Understand product usage in aggregate — how many users complete an arc, where drop-off happens	Anonymized + aggregated event data (via PostHog)
Communicate with you about your account or service updates	Email address
Comply with legal obligations and prevent abuse	Server logs, account info

05Sherp transcripts and PostHog

Because this is the most personal data flow in Inward, it gets its own section.

When you talk to Sherp — either inside an episode or in advisor chat — the full text of your message and Sherp's response is sent to PostHog, our analytics provider, as a "$ai_generation" event. PostHog stores it. We use it to:

See whether Sherp's quality is regressing across releases
Spot prompt-engineering bugs (e.g., Sherp ignoring your stated preferences)
Monitor cost per user and per call type
Reproduce specific bugs when you report them

PostHog is a third-party processor; we have a Data Processing Agreement with them. Their retention for these events is 1 year on our current plan, after which the events are deleted automatically. PostHog does not use your content for any purpose other than to provide service to us.

If you'd prefer your transcripts not be logged this way, the only option today is not to use Sherp. We expect to add an opt-out toggle for AI logging in a future release; until then, this is the trade-off the app makes to ship reliable quality.

06Subprocessors

Inward uses the following service providers ("subprocessors") to operate the app. Each receives only the data needed for its function. None of them sell your data.

Subprocessor	What it does	What it sees	Location
Anthropic (Claude API)	Generates Sherp's replies and most synthesis output	Your messages to Sherp + system prompts + AI responses	USA
OpenAI (when enabled)	Alternative LLM provider; routed for some calls when our internal AI-provider flag is set to OpenAI	Same as Anthropic, when used	USA
Supabase	Database, authentication, file storage	All persisted user data — account, responses, transcripts, Personal File, synthesis	USA (AWS regions)
Vercel	Hosting and serverless compute for the backend API	All request/response data (in-flight only — Vercel does not persist your content beyond logs)	USA
PostHog	Product analytics + LLM observability	Event data + AI prompt/completion content (see §5)	USA
Sentry	Error and crash monitoring	Stack traces, errors, request metadata (no transcript content)	USA
Apple	Sign in with Apple, App Store, push notifications (if you opt in)	Sign-in identity, subscription metadata, device push token	Per Apple

On AI training. Anthropic's API terms (anthropic.com/legal/commercial-terms) and OpenAI's API terms (openai.com/enterprise-privacy) both prohibit using API content to train their models. This is different from their consumer products (e.g., ChatGPT or claude.ai). Inward uses only the API path. We ourselves do not train models on user content.

On data location. All current subprocessors store data primarily in the United States. If you are in the EU, UK, or another region with cross-border transfer rules, see §10.

Changes. We may update subprocessors over time. Material changes are announced in-app, and the current list always lives at the latest version of this policy.

07Retention

Data type	Retention
Account and content (responses, conversations, Personal File, synthesis)	Kept while your account is active. Soft-deleted when you delete your account; hard-deleted from primary storage within 30 days of soft delete.
Database backups	Rotating, with a maximum age of 30 days. Hard-deleted content is purged from backups within 30 days.
AI events in PostHog (prompt + completion content)	1 year from event date, then auto-deleted
Aggregate analytics events in PostHog	1 year from event date
AI provider logs (Anthropic, OpenAI)	Up to 30 days for abuse monitoring per their API terms, then deleted
Server logs (Vercel)	30 days
Error reports (Sentry)	90 days

Account deletion is initiated from the app settings screen. We process deletion requests within 30 days. Some information may be retained longer if required by law (e.g., billing records).

08Your rights

8.1 Everyone

Regardless of where you live, you can:

See your data. Email privacy@getinward.com and we'll send you a copy of what we hold.
Correct your data. Request corrections to factual errors (e.g., a Personal File fact you disagree with).
Delete your account. Either from app settings or by emailing privacy@getinward.com.
Withdraw consent. Stop using the app at any time; this stops new data collection.

8.2 California residents (CCPA / CPRA)

If you are a California resident, you have additional rights:

Right to know what categories of personal information we collect, the sources, the purposes, and the categories of third parties we share with. This document discloses all of those.
Right to delete your personal information, subject to limited exceptions (e.g., legal compliance).
Right to correct inaccurate personal information.
Right to limit use of sensitive personal information. We treat your introspective content (Sherp transcripts, Personal File) as sensitive personal information. We use it only to provide the service to you — we do not use it to infer characteristics for advertising or for any other purpose.
Right to non-discrimination. We will not deny service or charge a different price for exercising any of these rights.
No "sale" or "share." We do not sell your personal information. We do not share it for cross-context behavioral advertising.

To exercise any of these rights, email privacy@getinward.com. We may need to verify your identity (typically by confirming your account email).

8.3 EU, UK, and other GDPR-aligned regions

If you are in the EU, UK, Switzerland, or another region with GDPR-style rules, you also have:

Right of access, rectification, erasure, restriction, portability, and objection under GDPR Articles 15–22.
Right to lodge a complaint with your local data protection authority.
Legal basis. We process your data under the contract basis (to provide the service you requested) and, where applicable, legitimate interests (e.g., security, abuse prevention, product improvement).
International transfers. Your data is stored in the United States. We rely on the EU–US Data Privacy Framework where available, and Standard Contractual Clauses with subprocessors as a backup.

09Children's privacy

Inward is rated for users 17 and older. We do not knowingly collect personal information from anyone under 13. If you are a parent or guardian and believe a child under 13 has provided us personal information, contact us at privacy@getinward.com and we will delete it.

10International users

Inward is operated from the United States. By using the app, you understand that your information will be processed in the United States and other countries where our subprocessors operate. Different countries have different privacy laws; we apply the protections in this policy globally.

11Security

We use industry-standard practices:

Encryption in transit. All app and API traffic uses TLS 1.2 or higher.
Encryption at rest. Supabase encrypts the database at rest.
Authentication. JWT-based session tokens with appropriate expiry. Sign in with Apple is supported.
Row-level access control. Our database enforces that each user can only read their own data, validated at the database engine level (Postgres Row Level Security).
Limited employee access. Only authorized engineers can access production systems, and only when needed for support or operations.

No system is perfect. If we discover a security incident affecting your data, we will notify you in line with applicable law.

12Changes to this policy

We update this policy as the product evolves. Material changes (new subprocessors, new categories of data, new uses) will be announced in-app with at least 30 days notice before they take effect. The "Effective" date at the top of this document reflects the most recent revision.

A change history is maintained in our public repository so you can see what changed and when.

13Contact

Get in touch

Inward Journeys, Inc.

61 Richardson St, Apt PHH, Brooklyn, NY 11211

privacy@getinward.com