GuardMe, Inc. ("GuardMe", "we") makes GuardMe, an app that protects you from scam, phishing and fraud websites. This policy says what we collect, why, how long we keep it, and what we never do.
The short version
- We don't sell or share your data with advertisers or data brokers, and our apps contain no advertising or third-party analytics code.
- We don't keep a history of the websites you visit. If you choose to help fix false alarms (Android and Windows only), we check the dangerous sites you open, with personal details removed and no device ID attached, and delete those records within 90 days.
- You don't need an account. We know your email address only if you subscribe on our website or email us.
- Our servers keep technical logs for up to 15 days, only to find and fix problems. They include IP addresses, and can show which blocked site an IP address asked for.
1. How protection works
- Checking sites. The app sends the name of each website you open to GuardMe's servers, which check it against our list of dangerous sites. We don't keep a record of which device looked up which site, except in support logs you choose to send us. If you choose to help fix false alarms, we also keep records of the dangerous pages you open (see below), but never with your device ID, account or IP address. Our technical logs can show which blocked site an IP address asked for, for up to 15 days. We do keep security records about websites themselves (for example, which addresses a website's name points to), with nothing about who looked them up.
- When our servers can't be reached. If our servers can't be reached, the app briefly uses your network's own name service (usually your router's), or on some devices Google's or Cloudflare's public one, so your internet keeps working. Whoever answers then sees the sites your device looks up and its IP address, under their own privacy policies.
- Sign-in pages and on-board services. On Wi-Fi with its own sign-in page or services, such as planes and hotels, the app also asks the network's own name service about the few names only it can answer: known in-flight and hotel Wi-Fi sign-in sites, and site names our servers say don't exist. That keeps sign-in and on-board entertainment working. Blocked sites stay blocked. The network sees those names.
- Warning pages and the GuardMe certificate. To show a warning page when you open a dangerous secure site, GuardMe needs you to install its certificate.
- Protection turns on only once it's installed. Without it, a blocked site would show an error with no way to continue. If you decline, GuardMe stays off until you finish setup.
- We use it only for sites on our danger list, and sites that route through them: for those, our server answers in the site's place and shows you the warning.
- If you choose to continue (except video-piracy sites, which GuardMe doesn't relay), our server opens the site for you and adds a reminder to its pages, until the site leaves the danger list. It counts how many pages are opened on that site each day. Unless you've chosen to help fix false alarms (below), it keeps nothing about you or what you do there. Your choice is stored only in your browser.
- We never use the certificate to open your connections to any other site.
- What we keep from warnings:
- daily counts per dangerous site: warnings shown, continued and declined, choices to continue that were later taken back, and pages opened after continuing. They contain no information about you, and we keep them as aggregate statistics with no time limit;
- the names of the sites that triggered a warning, without any information about who visited, for up to 90 days, so our systems can check those sites.
- Help fix false alarms (optional). On Android and Windows, you can choose to let us check the dangerous sites you open, so we can unblock safe sites sooner and spot fake stores. It's off unless you say yes, and you can turn it off anytime in Settings (section 6). The iPhone and Mac apps don't offer it.
- Which sites: only sites on our danger list, and sites that route through them. Never any other site.
- What we keep from your visit:
- how each page is built (its layout, scripts and forms), not the words or pictures you saw, and never what you type;
- the web addresses of the page and of its pictures;
- the size of the page and of each picture your browser received, and a short code calculated from each, never the content itself, so we can judge the site without keeping what you saw;
- the type of browser and device; how each page was opened (from a link, or as a picture or script inside another page, and whether from the same site or another one); and the names, never the values, of everything else your browser sent with the request.
- Nothing that says who you are: personal details are removed before anything is stored, and no device ID, account or IP address is attached. Pages are grouped only within one visit to one site in the same hour, and we don't link visits to different sites together.
- Who can see these records: only GuardMe's systems and the staff who review false alarms. Our own AI checks them on our own servers, and they're never sent to an outside AI company.
- How long: we delete these records within 90 days. What our checks conclude about a site, which contains nothing about you, we keep as long as we need it to judge that site.
- You'll see when it's on: the switch in the app's Settings shows it, and warning pages, and the reminder on sites you continue to, say that the page is being checked.
- Our own checks of dangerous sites. Our systems visit sites on the danger list themselves, using our own devices and connections, and keep copies of those public pages. This never uses your connection or anything about you.
2. What we collect
| What | When | What it contains | How long we keep it |
|---|---|---|---|
| Device certificate | When you set up the app | A random serial number and a GuardMe ID for your device, app and system version | 12 months after your device was last seen |
| App statistics | On Android and Windows unless you turn it off in Settings; on iPhone and Mac only if you choose "Share" | How far setup got, app and system version, installs, updates and uninstalls, and your device's GuardMe ID (on Android and Windows, the ID the system provides) | 12 months |
| Support logs | Only when you tap Report an Issue or Send Support Logs | The app's recent activity log, including the names of blocked sites looked up recently, device ID and versions, and anything you type. Never the content of your traffic. | 90 days |
| Help fix false alarms | Only if you choose it, on Android and Windows, when you open a site on our danger list or one that routes through them | How those pages are built (not the words or pictures you saw), the addresses of the page and its pictures, the size of what your browser received with a short code calculated from it, and the type of browser and device, and how each page was opened. Personal details removed; no device ID, account or IP address. | 90 days; what our checks conclude about the sites, with nothing about you, as long as we need it |
| Subscription | If you subscribe on our website | Your email address, your device's GuardMe ID and certificate serial, plan and status, and Stripe's customer ID. We never receive your full card number. | While your subscription is active, then 24 months |
| Full-device VPN (subscription) | While you use it | When you connected and disconnected, the region, how much data was sent and received, your VPN key, and the address our VPN gives your device inside it, linked to your device's GuardMe ID and certificate serial so your subscription reaches the right device. No IP address of yours, no sites, no traffic content. | 90 days after you disconnect; an unused VPN key is removed after 90 days |
| Emails we send you | Receipts and subscription notices | Your email address, and what the email says | Sent through our email provider; see section 3 |
| Emails you send us | If you email us | Your email address, and what you write | In our mailbox (see section 3) until you ask us to delete them |
| Do-not-email list | If an email to you bounces or you mark it as spam | Your email address and the reason | Kept, so we never email you again |
| Technical logs | While our servers run | IP addresses and technical details of requests, never email addresses. They can show which blocked site an IP address asked for. | Up to 15 days |
We don't collect your name, phone number, postal address, location, contacts, photos, or the content of your messages or browsing. If you choose to help fix false alarms, we keep how the dangerous pages you open are built (section 1).
3. Who else handles data
Companies that handle data about you:
- Amazon Web Services hosts our servers and storage.
- Stripe processes subscription payments. It collects your email and payment details itself, under Stripe's privacy policy. We pass it your device's GuardMe ID, certificate serial and plan details so your subscription reaches the right device.
- Resend sends our emails, so it receives your email address and the email's content.
- Google Workspace hosts our email, so it holds any email you send us and our replies.
- Anthropic makes the AI assistant our staff use to read support logs and server logs when fixing a problem, so it processes what those logs contain.
- Affiliate partners. If you tap one of the safer alternatives on a warning page, you go to that partner's site, which may know you arrived from GuardMe. We don't send them anything about you.
- Apple, Google and Microsoft distribute our apps under their own policies.
Services that see only the names of sites. We send them site names and nothing about who asked:
- Google's public name service, which our servers use to look up the addresses of dangerous sites.
- Domain registries, which our systems ask about the sites they check.
- Sites' own name servers, which our servers ask for the addresses of sites.
If our servers can't be reached, the app briefly uses your network's own name service (usually your router's), or on some devices Google's or Cloudflare's public one, so your internet keeps working. Whoever answers then sees the sites your device looks up and its IP address, under their own privacy policies.
On Wi-Fi with its own sign-in page or services, such as planes and hotels, the app also asks the network's own name service about known in-flight and hotel Wi-Fi sign-in sites, and about site names our servers say don't exist, so sign-in and on-board entertainment keep working. The network sees those names.
That's the complete list. We don't sell data, and we don't share it with advertisers or data brokers. We don't sell, use or disclose your data for anything else; the companies above handle it only to run GuardMe.
If you choose to help fix false alarms, what we keep is checked by our own AI on our own servers (hosted by Amazon Web Services, like the rest of GuardMe). It's never sent to Anthropic or any other outside AI company.
4. Our website
guardme.online uses no analytics, advertising or tracking cookies, and loads nothing from other companies. Warning pages store your Continue choice in your own browser only.
5. Keeping data safe
- Connections between the app and our servers are encrypted, with two exceptions: if encrypted lookups can't get through, the app may check site names with our servers without encryption; and warning pages for sites that don't use encryption themselves aren't encrypted either.
- Access to stored data is limited to what each part of our system needs. Records kept to help fix false alarms are stored apart from everything else. Only GuardMe's systems and the staff who review false alarms can see them, and each time staff open one, it's logged.
- Server backups are kept for up to 8 weeks.
- All our server logs are kept for up to 15 days and used only to find and fix problems. They never contain email addresses.
6. Your choices and rights
- Turn statistics on or off anytime in the app's Settings.
- Turn Help fix false alarms on or off anytime in the app's Settings, on Android and Windows. Turning it off stops it within a few minutes. The records already kept carry no device ID, so we can't pick out yours to delete on request; they're deleted automatically within 90 days.
- See your device ID in the app's Settings.
- Delete your data: email privacy@guardme.online with your subscription email or your device ID. We'll delete everything linked to it within 30 days, apart from anything we must keep by law, such as payment records held by Stripe. If your subscription is still active, we cancel it first. Copies in our server backups expire within 8 weeks.
- Uninstalling stops all collection, apart from one last statistic on Windows and Mac saying the app was uninstalled (only if statistics are on). On iPhone and Android, the GuardMe certificate stays installed until you remove it in Settings. Data already collected is deleted on the schedule above, or sooner if you ask.
These rights apply wherever you live, including under GDPR and CCPA.
7. Children
GuardMe isn't directed at children under 13, and we don't knowingly collect their personal information. If you believe we have, contact us and we'll delete it.
8. Changes
We'll post changes here and tell you in the app before they take effect. If a change would collect something new, we'll ask first.
9. Contact
privacy@guardme.online. GuardMe, Inc., Idaho, USA.