Safe to Tap
Privacy Policy
Effective October 30, 2026. Safe to Tap was called Scam Lens until October 2026; nothing about how we handle data changed with the name.
- Apart from the optional Scam Text Filter, we only see what you choose to check. We analyze it, return a verdict, and don't keep it.
- If you turn on the Scam Text Filter, texts from numbers that aren't in your contacts are checked automatically as they arrive. They aren't stored or linked to you.
- Your scan history stays on your iPhone.
- Screenshots are read on your iPhone. Only the extracted text is sent, never the image.
- We don't ask for your name or email, show ads, track you, or sell data.
- You can delete your account in the app at any time, which deletes everything we hold.
What happens when you check something
When you paste or share a message, email, link or screenshot into Safe to Tap and tap check:
- For screenshots, your iPhone reads the text itself. The image never leaves your device.
- The text (and any details the app knows, such as the sender, subject, or which app you shared from) is sent over an encrypted connection to the Safe to Tap server.
- Our server checks any links it finds (see “Link checks” below) and sends the text, together with the link check results, to our AI provider, Anthropic, for analysis. Your account details are not sent to Anthropic.
- The verdict comes back to your iPhone and is saved in your on-device history.
We ask for your permission in the app before the first check, and you can withdraw it at any time in the app's Account screen. Without it, Safe to Tap can't analyze anything.
Scam Text Filter (optional)
The filter is off until you allow it in the app (Account → Scam Text Filter) and choose Safe to Tap in your iPhone's Settings (Apps → Messages → Unknown Senders → Text Message Filter). When it's on:
- When a text arrives from a number that isn't in your contacts, your iPhone sends the text and the sender's number to the Safe to Tap server. Only regular text messages (SMS and MMS) are checked, not iMessages. Texts from your contacts are never sent, and Safe to Tap on your iPhone skips verification codes.
- Our server checks any links and asks our AI provider, Anthropic, whether it's a scam. The request carries no account information, so we can't tell whose phone it came from.
- If it's clearly a scam, your iPhone moves it to the Junk folder in Messages. Otherwise it arrives as usual.
We don't store filtered texts, their senders, or the result. Turn the filter off at any time in the app (Account → Scam Text Filter → Stop Checking Texts), or by choosing None under Text Message Filter in Settings.
What we store
| Data | Why | How long |
|---|---|---|
| An account ID, and the anonymous identifier Sign in with Apple gives us for this app | To recognise your account and apply your plan's limits | Until you delete your account |
| An encrypted token from Apple | Only to disconnect Safe to Tap from your Apple Account when you delete your account | Until you delete your account |
| For each check: a one-way fingerprint (hash) of the text, the type of item, the score and verdict, which AI model produced it, the time, the website names of any links sent for a sandbox check, and your answer if you tell us whether the verdict was right | To count your daily checks, to investigate abuse of the service, and to measure and improve accuracy | Deleted automatically after 30 days, or immediately when you delete your account |
| Server logs with IP addresses and request times | Security and keeping the service running | A short period, then deleted by routine log rotation |
We do not store the text you check. A fingerprint can't be turned back into the text.
Who else is involved
| Provider | What they receive | Why |
|---|---|---|
| Anthropic (Claude AI) | The text you check and the link check results. With the Scam Text Filter on, also texts from unknown numbers and their sender's number. No account details. | To analyze the item for signs of a scam. Under Anthropic's commercial terms, this data isn't used to train its models. Anthropic may keep it for a limited period (currently up to 30 days) for safety and abuse monitoring. See Anthropic's privacy policy. |
| Cloudflare (URL Scanner) | Up to three links from the item you check | Cloudflare opens the website in an isolated browser and reports whether it's malicious. Links are submitted as unlisted, so they don't appear in Cloudflare's public scan feed. See Cloudflare's privacy policy. |
| Domain registries (via rdap.org) | Only website names, such as example.com | To look up how old a website is. New websites are a common scam sign. |
| Apple | Sign-in requests | To sign you in with your Apple Account. We don't request your name or email. |
We don't sell or rent any data, use it for advertising, or share it with anyone else, except where the law requires it.
On your iPhone
Your scan history, including a copy of each item you checked (for screenshots, a smaller copy of the image), is stored only on your iPhone, inside the Safe to Tap app. You can delete individual entries or the whole history in the app. Deleting the app removes it too.
Your choices and rights
- Delete your account: Account → Delete Account. This deletes your account and all records we hold about you, disconnects Sign in with Apple, clears your on-device history, and turns off the Scam Text Filter.
- Withdraw AI consent: Account → AI Analysis. Safe to Tap stops sending anything for analysis, including texts from the Scam Text Filter, until you agree again.
- Turn off the Scam Text Filter: Account → Scam Text Filter → Stop Checking Texts, or choose None under Text Message Filter in iOS Settings.
- Stop using Sign in with Apple: in iOS Settings → your name → Sign in with Apple. We then delete your account automatically.
- Questions or requests about your data (including access or deletion under laws such as the GDPR): contact us at the address below.
Children
Safe to Tap isn't designed for children under 13, and we don't knowingly collect data from them.
Security
All communication between the app and our servers is encrypted. Access to our servers is restricted, and the Apple token we keep is encrypted at rest.
Changes
If we change how Safe to Tap handles data, we'll update this page and its effective date. Significant changes will also be shown in the app.
Contact
Email: privacy@prodevy.com