Safety is not a feature.
It is the foundation.
Heard was built from day one with your safety in mind. Here is how we protect you, what we monitor, and what we report.
How we protect you
No photos until you both agree
Photos are locked behind a mutual consent system. Neither person sees the other's photos until both explicitly agree to share. This is not optional, it is how the app works.
Voice-first by design
Your voice intro is how people meet you. This makes it significantly harder to create fake accounts. A voice is harder to fabricate than a stolen photo.
Date safety check-ins
When you plan a date through Heard, you can enable real-time safety check-ins. The app prompts you to confirm you are safe at intervals you set. If you do not respond, your emergency contact is alerted with your last known location.
Report and block within 2 taps
Every conversation and profile has both report and block accessible within 2 taps. Blocking is immediate and permanent. Reports are reviewed by our team.
Row-level security on every table
Your data is protected at the database level. Not just by application code, but by security policies on every single table. A user can only access their own data. This is enforced by the database itself, not just the app.
Your data is encrypted
All data transmitted between your device and our infrastructure is encrypted in transit using TLS (the same standard used by banks). Your data is also encrypted at rest by our database provider, Supabase, which runs on AWS with AES-256 disk encryption. No one can intercept your messages while they travel, and the underlying storage cannot be read without authorized access.
We do not use end-to-end encryption for messages. Here is why: E2E encryption would prevent us from reviewing reported content, detecting illegal material, and complying with legal obligations. Every major dating app (Hinge, Bumble, Tinder) makes the same decision for the same reasons.
Your messages are protected by access control at the database level. Security policies ensure that other users cannot access your conversations. Only you and the person you are matched with can read your messages through the app.
Our team does not read your messages unless there is a reason to. We have administrative access to the database for operational purposes, but we do not browse conversations. When a user files a report, that report is what triggers a review of the flagged conversation. We may also access message data to comply with legal requests, investigate credible safety threats, or enforce our terms of service. We are transparent about this because we believe honesty is better than a false promise of invisibility.
Content moderation
Heard uses both automated and human moderation to keep the platform safe.
Automated scanning
- Every photo uploaded to Heard is automatically scanned using Microsoft PhotoDNA, the industry standard for detecting child sexual abuse material (CSAM). This happens before any photo becomes visible to other users. Flagged content is immediately hidden and the account is suspended pending review.
- Text messages are scored for toxicity, threats, and sexually explicit content using Google's Perspective API. Messages above safety thresholds are blocked or flagged automatically.
Human review
When a user reports content, our team reviews it manually. Here is what we can review:
- Voice intros and voice notes can be listened to when reported for harassment, threats, or inappropriate content
- Photos shared through the album system can be reviewed when reported for policy violations
- Text messages in a reported conversation are reviewed to assess the report
- Bios and profile content can be reviewed and removed if reported
When a violation is confirmed, we take action ranging from a warning to permanent suspension depending on severity. Users are notified of the outcome through their Trust Center in the app.
What we report to authorities
Heard has a zero tolerance policy for the following. We will report to the appropriate authorities without exception:
Child Sexual Abuse Material (CSAM)
NullDaily LLC has a zero-tolerance policy for CSAM. Every photo uploaded to Heard is automatically scanned using Microsoft PhotoDNA before it becomes visible to other users. Any content identified as CSAM results in immediate account suspension and mandatory reporting to NCMEC via the CyberTipline as required by federal law (18 U.S.C. 2258A). NullDaily LLC is a registered Electronic Service Provider with NCMEC. Flagged content is preserved as evidence until NCMEC confirms receipt of the report.
Threats of violence
Credible threats of violence against any person are reported to local law enforcement. If there is an imminent threat, we contact emergency services immediately.
Human trafficking
Any patterns indicating trafficking or exploitation are reported to the National Human Trafficking Hotline and relevant law enforcement.
Self-harm and suicide
If we become aware of an imminent risk of self-harm, we may contact emergency services and provide relevant information to help protect the individual.
We comply with valid legal process including subpoenas, court orders, and search warrants from law enforcement agencies. We will notify affected users unless legally prohibited from doing so.
Account standing
Every Heard account has a standing level visible in your Trust Center:
No violations. Full access to all features.
Minor violation. Some features restricted.
Multiple violations. Account under review.
Serious violation. Account access removed.
Violations are reviewed individually. Standing is visible to you at all times so there are no surprises.
You are in control
- Delete your account at any time from Settings. We process deletions within 7 days and remove all your data, including voice recordings, photos, messages, and profile information
- Block any user instantly. They will not be able to see your profile or contact you
- Control your visibility with privacy toggles for age, distance, identity, Heard Score, game stats, active status, and read receipts
- Sleep mode hides your profile completely for up to 7 days while preserving your data and streaks
- Notification controls for every category: matches, messages, streaks, daily prompts, dates, safety, and marketing
- Photo sharing is always mutual. You choose when and with whom you share your photos
Transparency reports
We are committed to publishing quarterly transparency reports. These will include:
- How many reports were filed and how many resulted in action
- How many times administrative database access was used and for what purpose
- How many legal requests we received and how many we complied with
- How many accounts were suspended or deleted for violations
- How many photos were scanned by PhotoDNA and how many were flagged
- Any content reported to NCMEC or law enforcement (count only, no identifying details)
- Any changes to our moderation policies or security practices
When we say our team does not read your messages unless there is a reason to, transparency reports are how we prove it. We will audit our own access and publish the results. If we accessed data, you will know how many times, why, and what category it fell under.
The first transparency report will be published after our public launch.
Questions or concerns?
If you have a safety concern or need to report something urgently, contact us directly.
safety@heardapp.co