Relationship-message privacy must be explicit, not implied.
Grey Mirror handles sensitive relationship-message uploads. This page keeps the public guidance conservative and routes account-specific requests to the formal privacy, data-deletion, trust, and contact surfaces.
How should users think about privacy?
Users should treat relationship-message uploads as sensitive personal data. Grey Mirror public copy avoids unsupported security, compliance, retention, deletion, or training-data claims and links to current policy surfaces for account-specific requests.
What data may be involved?
A Grey Mirror run may involve raw uploads, normalized messages, participant labels, parser diagnostics, report artifacts, cached assets, account identifiers, and payment or unlock state depending on the flow.
Which details belong on the formal policy pages?
The public explainer should not turn operational details into promises unless the formal policy pages also say them. Use the linked pages for current account, deletion, access, and privacy-request language.
- Raw upload and normalized-data retention windows belong on the current privacy and deletion pages.
- Report retention, cache cleanup, backup expiry, and temporary export cleanup should match the implemented policy surface.
- Deletion, export, and access-request workflows should route through the site’s account, privacy, deletion, trust, or contact paths.
- Training-data and subprocessor language should stay conservative unless the current policy states a specific claim.
- Region-specific rights language should appear in the formal policy pages, not in isolated product copy.
What should the product never claim without proof?
Trust copy is only useful if it is accurate. Grey Mirror should not claim certifications, encryption specifics, compliance status, region-specific rights handling, or subprocessor lists unless the underlying policy and implementation support it.
- No fabricated SOC 2, HIPAA, GDPR, or ISO claims.
- No unsupported encryption-at-rest or end-to-end encryption language.
- No fake deletion SLA.
- No unsupported training-data promise.
Where should users go next?
Users should read the formal privacy policy and contact support for account-specific requests. This page is a product-specific privacy explainer, not a replacement for approved legal policy.
Frequently Asked Questions
Can I delete my uploaded conversation?
Use the data-deletion, privacy, trust, or contact routes for the current request path. Grey Mirror should not promise a deletion outcome in a guide page that the formal policy does not support.
Does Grey Mirror train on my messages?
This guide avoids unsupported training-data claims. Check the current privacy and trust pages for the live policy language.
Is Grey Mirror HIPAA or SOC 2 compliant?
No such claim is made here. Public copy should not state compliance certifications unless they are verified and approved.
References and methodology
Related Grey Mirror guides
- Do slower replies mean less?
- How to judge an analyzer
- Analyze before you reply
- Relationship Texting Benchmarks 2026
- Relationship text analyzer
- Can AI analyze a full text message history?
- Methodology
- Public white paper
- Metrics library
- Evidence standards
- Privacy and deletion
- AI sycophancy vs measured analysis
- Instagram DM analyzer
- Couples text message analyzer
- Interactive sample report
- Relationship text analysis glossary
- Long-term pattern analysis
- Love language in texting
- iMessage analysis
- WhatsApp chat analysis
- ChatGPT vs Grey Mirror
- Screenshots vs full thread
- Repair attempts in texting
- Conflict escalation patterns
- Emotion word frequency
- Texting anxiety signs
- Friendship text analysis
- Telegram text analysis
- Improve text communication
- Apology insufficiency case study
- Analyze chat history for patterns
- SMS and Android text analysis
- Pricing and free preview
- Private relationship text analyzer
- Best AI text message analyzer
- Best text message analyzers 2026
- Chat analyzer comparison
- Red flag text analyzer
- Situationship text analyzer
- Relationship pattern analysis case study
- Criticism in texts case study
- Dismissiveness case study
- Emotional availability case study
- Emotional labor case study
- Emotional tone drift case study
- Talking about problems case study
- Mixed signals case study
- Post-conflict patterns case study
- Power dynamics case study
- Reading subtext case study
- Validation in texts case study
- Grey Mirror blog
Start here
- Analyze your messages — upload a full exported history on the web and watch the first scenes free.
- Grey Mirror for iPhone — the free native iOS app, App Store id 6799236359, iOS 18 or later. Same exports, same report as the web.
- Pricing — the complete report is a one-time $25 unlock, no subscription.
- Published research — aggregate benchmarks across 4,600,611 messages, with the withheld measures stated.
- The Signal Is Rare — conflict outran repair 36 to 1, 97.4% of messages carried no detectable signal, and one claim was withheld.
More from Grey Mirror
View the canonical Relationship-message privacy must be explicit, not implied. page