1. Controller and contact
The controller for processing on onlinebetrug.ai within the meaning of the General Data Protection Regulation (GDPR) is: OnlineBetrug AI, a product of QAD Soft, Dennis Bölling, Fichtelgebirgestraße 14, 78141 Schönwald im Schwarzwald, Germany.
E-mail for any data protection question and to exercise your rights: hello@onlinebetrug.ai.
We are not required to appoint a data protection officer under Article 37 GDPR.
onlinebetrug.ai and fakeshop.ai are run by the same controller on a shared server with a shared database. This policy covers onlinebetrug.ai. For fakeshop.ai, its own privacy policy applies.
2. Overview
On onlinebetrug.ai you can check e-mails, messages, links, phone numbers, IBANs and sender addresses, compare a password against published password lists and check your e-mail address for known data breaches. None of these checks needs an account. We use what you enter for the check and to prevent misuse. We do not use it for advertising, we do not build profiles of you and we do not sell data.
For each processing activity the sections below set out which data we process, why, on what legal basis, who receives it, whether it leaves the EU and when we delete it. Many checks involve data about other people, such as the sender of a scam e-mail. More on this under third-party data.
Please send us only what the check needs. Before you do, remove details about health, religion, political opinions or other special categories of data (Article 9 GDPR), your login details and one-time codes. We do not want to process such data. If it is nevertheless contained in an e-mail or message you submit, it goes through the automated check and is deleted with the rest of the content within the periods given below.
3. Visiting the website, server logs and hosting
When you open onlinebetrug.ai, your browser sends technically necessary information to our servers. The request first reaches a front server of our own (reverse proxy), which accepts the encrypted connection, and is then passed to the application web server. Both keep access logs.
- Data
- IP address, date and time, requested address including path and parameters, status code, amount of data transferred, previous page (referrer), browser identifier (user agent).
- Purpose
- Delivering the website, troubleshooting, detecting and fending off attacks.
- Legal basis
- Article 6(1)(f) GDPR. Our legitimate interest is secure and reliable operation.
- Recipients
- Hetzner Online GmbH, Industriestr. 25, 91710 Gunzenhausen, Germany, as our processor (Article 28 GDPR) running our servers, database and mail servers in data centres in the EU.
- Third country
- No.
- Retention
- As long as needed for troubleshooting and fending off attacks. We will state a fixed maximum period here once it has been configured on the servers.
The application web server leaves out of its log the addresses where a checked value or an access key may appear in the request: the report pages of the e-mail check (/mail-check/b/…), fetching a report, the IBAN check, the password check, the sender address check and the breach check. It logs all other requests, including the phone number check (the number is in the body of the request, not in the address) and searches in the warnings (the search term is in the address). The front server logs all requests.
4. Audience measurement with Rybbit
We measure the use of this website with the open-source software Rybbit. We run Rybbit ourselves on our own servers in the data centre of Hetzner Online GmbH in Germany. No data is transmitted to third parties or to countries outside the EU. Rybbit sets no cookies and stores nothing on your device. We do not embed advertising or tracking services from other providers.
- Data
- Page viewed, page title, referring source, time, screen size, browser language, browser, operating system, device type, approximate location (country, region, city), technical errors in the browser and page load times. We also record selected interactions, such as the start of a check or a click on an important button, without any content and without any link to you. Your IP address is not stored. It is used only briefly in memory to determine the approximate location and, together with the browser identifier, to form a pseudonymous identifier. That identifier is formed with a random value that changes daily, so visits on different days cannot be linked. Parameters in addresses that may contain credentials or identifiers (for example token, code, email) are removed before storage. The report pages of the e-mail check and the result page of the breach check, whose address carries a key, are not measured at all.
- Purpose
- Knowing which pages and checks are used, so we can improve the service.
- Legal basis
- Article 6(1)(f) GDPR. Our legitimate interest is a low-data audience measurement to develop the service.
- Recipients
- None apart from our host (section 3).
- Third country
- No.
- Retention
- At most 24 months, after which we delete the data.
Objection: if your browser sends the “Do Not Track” or “Global Privacy Control” signal, no measurement takes place. You can also switch the measurement off in this browser here. At your request this stores a setting on your device. You can also object by e-mail (section 21).
5. Protection against misuse: limits and CAPTCHA
To stop anyone using the checks in bulk, filling other people’s inboxes with e-mails or trying out numbers, we limit the number of requests.
Rate limiting
- Data
- Your IP address and the number of your requests in the current time window.
- Purpose
- Preventing overload and automated use.
- Legal basis
- Article 6(1)(f) GDPR. Our legitimate interest is a stable service free of misuse.
- Retention
- In the memory or the cache (Redis) of our server, for a time window of one minute to one hour depending on the check; then the entry expires.
Daily counters
- Data
- For the daily limits we do not store your IP address. We derive a keyed value (HMAC) from it using a key that changes every day; for IPv6 from the network range (/64 and /48). Plus the number of checks on that day.
- Purpose
- A daily limit per connection for checks that use a language model, the sender address check and the password check.
- Legal basis
- Article 6(1)(f) GDPR (as above).
- Retention
- We delete the counters seven days after the day counted.
CAPTCHA (Cloudflare Turnstile)
- Data
- When you check an e-mail, message, phone number or sender address and for the breach check, the page loads Cloudflare’s Turnstile script. It receives your IP address and characteristics of your browser and device. Our server sends the result together with your IP address to Cloudflare for confirmation.
- Purpose
- Telling whether a person or a program is requesting the check.
- Legal basis
- Article 6(1)(f) GDPR. Our legitimate interest is protection against automated misuse, for the breach check in particular protecting other people’s inboxes. For reading characteristics of your browser we rely on section 25(2) no. 2 of the German TDDDG, because it is strictly necessary for the check you requested.
- Recipients
- Cloudflare, Inc., 101 Townsend St., San Francisco, CA 94107, USA.
- Third country
- USA. We base the transfer on the EU-U.S. Data Privacy Framework (adequacy decision of the European Commission of 10 July 2023), to the extent Cloudflare is certified under it, and on the European Commission’s standard contractual clauses (Article 46(2)(c) GDPR).
- Retention
- We do not store the result. How long Cloudflare keeps data is governed by Cloudflare’s privacy policy.
The CAPTCHA is only active if it has been set up on the server. We do not offer the breach check without a CAPTCHA.
6. Checking an e-mail (disposable address, forwarding, upload)
You can forward a suspicious e-mail to a disposable address or upload the e-mail file (.eml). Our system analyses it automatically and produces a report. If you forward the e-mail, we also send you the report by e-mail.
Process and data
- Disposable address: we generate a random address (pruefen-…@check.onlinebetrug.ai). It accepts up to three e-mails for 24 hours. We do not ask for your own e-mail address for this. The page keeps the address and an access key in your browser so it can show you the report (section 19).
- Receipt: our own mail server accepts forwarded e-mails. Our system collects them there and then deletes them from the mailbox.
- Raw copy: we store the complete e-mail with headers, text, links and attachments encrypted (AES-256-GCM) until it has been analysed. We delete it after the analysis, at the latest after 24 hours.
- Analysis: we check the sender and delivery path (SPF, DKIM, DMARC), links, phone numbers, IBANs, crypto addresses and attachments. We do not open attachments; we read only name, type, size and a checksum (SHA-256). Subject and text go to a language model after redaction (section 13). For lookups outside our systems see section 14.
- Report: it contains the result, the findings, the subject, display name and address of the sender of the checked e-mail, the links in defanged form, details of attachments and the phone numbers, IBANs (shortened) and crypto addresses found. It does not contain the raw copy or the e-mail’s HTML. We redact your addresses and your name in the subject and text, as far as we can recognise them from the recipient details.
- Reply e-mail: if you forward an e-mail, we send the report to the address you forwarded from. We only reply if your mail provider has signed the forwarded e-mail cryptographically (DKIM) and it is no more than two hours old. This way nobody can have reports sent to other people’s addresses. To open the report we store your address only as a keyed value (HMAC). We use it to check the address you enter the first time you open the report, and to limit replies to ten per address and day.
- Purpose
- The check you requested and the report.
- Legal basis
- Article 6(1)(b) GDPR for your own data (providing the check you asked for). For data about other people contained in the e-mail, Article 6(1)(f) GDPR: your legitimate interest and ours is to detect and prevent fraud.
- Recipients
- Hetzner (hosting, section 3), the provider of the language model (section 13) and the parties listed in section 14 with the data stated there. We send the reply e-mail through our own mail server; your mail provider receives it as recipient.
- Third country
- Only for the language model, see section 13.
- Retention
- Disposable address 24 hours. Raw copy until analysed, at most 24 hours. Report including the keyed value of your address 7 days. Characteristics of the perpetrators 180 days after the last sighting (section 15).
Only you and we know the link to the report; search engines are blocked from the report page. If you want a report deleted before the seven days are up, write to us with the link (section 21).
7. Checking a message or text message
You can paste the text of a text message or other message and give the sender. Alternatively you can choose a screenshot. Text recognition then runs in your browser: the image does not leave your device. You see the recognised text and can change it before sending. Only the text you send reaches us.
- Data
- Sender (at most 100 characters) and text (at most 2,000 characters).
- Process
- We store text and sender encrypted until they have been analysed and delete them afterwards, at the latest after 24 hours. The analysis works as for the e-mail check (section 6): links, phone numbers, IBANs, comparison with lists (section 14) and a language model (section 13). Before the language model we replace e-mail addresses in the text with a placeholder, except that of the sender. We do not redact names or other details in the text here.
- Report
- Sender, result, findings with short quotes from the text, links in defanged form, phone numbers, IBANs (shortened) and crypto addresses found. Available for 7 days through a key that only your browser holds.
- Purpose and legal basis
- As in section 6.
- Retention
- Raw text at most 24 hours, report 7 days, characteristics of the perpetrators 180 days (section 15).
Our programming interface (POST /api/message-check) can also check a text without a report. There the text goes to the language model without redaction and is neither stored nor logged. For metering quotas we store, for requests without an API key, the IP address with the number of requests per day for 35 days.
8. Checking a phone number
- Data
- The phone number you enter. It usually belongs to someone else.
- Process
- We classify the number with a software library and the numbering plans of the German Federal Network Agency (Bundesnetzagentur) and the Austrian regulator RTR, and compare it with the Bundesnetzagentur’s list of enforcement measures and with verified reports in our own data. All lists are held by us; the number is not sent to any other service.
- Purpose
- The check you requested.
- Legal basis
- Article 6(1)(b) GDPR; for the number as data about a third party, Article 6(1)(f) GDPR (protection against fraud).
- Recipients
- None apart from our host.
- Retention
- We do not store the number and do not write it to any log. The number is in the body of the request; the access log contains only the address of the interface (section 3).
9. Checking an IBAN
- Data
- The IBAN you enter. It usually belongs to someone else and may therefore be personal data.
- Process
- Your browser checks the check digits first. The IBAN goes to our server in the body of the request, never in the address. We check length and check digits, identify the bank using the Deutsche Bundesbank’s bank code file and the register of the European Banking Authority (both held by us) and count verified reports from people affected that name this IBAN. We only show you the number of reports and the date of the first one, never who reported it or the account holder.
- Purpose
- The check you requested.
- Legal basis
- Article 6(1)(b) GDPR; for the IBAN as data about a third party, Article 6(1)(f) GDPR (protection against fraud).
- Recipients
- None apart from our host.
- Retention
- We do not store the IBAN and do not write it to any log; the address of the IBAN check is excluded from the access log.
The reports we count come from people who reported fraud through fakeshop.ai, and we review them before they are added. How we process reports is set out in the privacy policy of fakeshop.ai.
10. Assessing a sender address and checking a link
- Data
- The e-mail address you enter. It is usually the address of a sender, so of another person.
- Process
- We check the spelling and similarity to known brands, query the Domain Name System (DNS) for the domain’s records, ask the responsible registry (RDAP or WHOIS) for the domain’s registration date and compare the domain with lists (section 14). Only the domain, the part after the @, leaves our systems. We do not contact the sender’s mail server. We compare the full address with our own findings from the e-mail check using its keyed value (HMAC).
- Purpose
- The assessment you requested.
- Legal basis
- Article 6(1)(f) GDPR. Your legitimate interest and ours is protection against fraud.
- Recipients
- For the domain: DNS servers and registries (section 14).
- Retention
- We do not store or log the address. We keep results per domain for up to six hours so that we do not query the same domain repeatedly.
Checking a link
- Data
- The link or domain you enter. Only the domain is processed; path, parameters and fragment are discarded. The domain usually belongs to a third party.
- Process
- We compare the domain with our own data sets: the Phishing.Database, the CERT.PL warning list (only if switched on), the warnings of the regulators (BaFin, AMF) and our records at fakeshop.ai with verdict and date of the check (section 14). Rules about how the address is written are added (short-link service, special characters, brand name in an unrelated domain, IP address). We do not open the link, do not query DNS or registries and do not start a check at fakeshop.ai. The link in the result view to the full shop check takes effect only when you click it; then the statements of fakeshop.ai apply.
- Purpose
- The assessment you requested.
- Legal basis
- Article 6(1)(b) GDPR; for the domain as a third party’s data, Article 6(1)(f) GDPR. Your legitimate interest and ours is protection against fraud.
- Recipients
- None apart from our hosting provider (section 3).
- Retention
- We do not store the address and do not write it to any log. The address is in the body of the request, never in the URL; the access log contains only the path of the interface. Section 5 applies to rate limiting.
11. Password check
The password check shows whether a password appears in published collections of stolen passwords. We use the “Pwned Passwords” service of Have I Been Pwned for this.
- Process
- Your password does not leave your browser. The browser computes a checksum (SHA-1) and sends only its first five characters to our server. Our server fetches from Pwned Passwords the list of all known checksums beginning with these five characters; the list contains several hundred entries each time. Your browser itself looks for the rest of its checksum in it. We learn neither the password nor the result.
- Data held by us
- The five characters and your IP address, for rate limiting and the daily counter (section 5). We do not write the five characters to any log; the address is excluded from the access log. We keep the Pwned Passwords response per five characters in memory for up to six hours, with no link to you.
- Purpose
- The check you requested.
- Legal basis
- Article 6(1)(b) GDPR; for protection against misuse, Article 6(1)(f) GDPR.
- Recipients
- Have I Been Pwned, operated by Superlative Enterprises Pty Ltd, Queensland, Australia; delivered through Cloudflare. The service receives the five characters and our server’s IP address, not yours.
- Third country
- Australia. There is no adequacy decision of the European Commission for Australia. We send nothing that Have I Been Pwned could link to you: not your IP address, not the password and not the full checksum.
- Retention
- As above; the page stores nothing on your device.
We do not call the method “anonymous”: the five characters narrow down the set of possible passwords. Only check a password you still use if you trust this method, and change it if it is found.
12. Breach check (your own e-mail address)
The breach check tells you in which known data breaches your e-mail address appears. So that only the owner of the mailbox sees the result, it works through a confirmation e-mail.
Process
- You enter your e-mail address and complete the CAPTCHA (section 5). Without a CAPTCHA set up, the breach check is switched off.
- We check whether mail can be delivered to this address (DNS lookup of the domain) and send, through our own mail server, an e-mail with a link. The e-mail contains no result.
- Until you open the link we store your address encrypted (AES-256-GCM), for at most 30 minutes. We store the key in the link only as a keyed value. When you open the link, your browser sends the key to us in the body of the request; the address is then deleted from the database. The link works once. If no source answers, the link stays valid until the 30 minutes are up so that you can try again.
- We then send LeakCheck the first 24 characters of a checksum (SHA-256) of your address, not the address itself. We get back the names of the breaches, their date and the types of data affected, no passwords and no records. We add month and data types from a directory of breaches that we build from the open list of Have I Been Pwned; nothing about you is sent for this.
- We show the result only to you. We do not store it, neither in the database nor in a cache, and there is no audience measurement on the result page.
- So that nobody uses us to fill other people’s inboxes, we count requests per mailbox and day (at most two e-mails per mailbox and day, currently at most 100 in total per day). For this we store a keyed value (HMAC) of the address, computed together with the date, and delete it two days after the day counted.
- Purpose
- The check you requested; protecting other people’s inboxes from misuse.
- Legal basis
- Article 6(1)(b) GDPR for the check, Article 6(1)(f) GDPR for protection against misuse.
- Recipients
- LeakCheck (leakcheck.io), based in London, United Kingdom. LeakCheck states that it does not store search queries. The shortened checksum is pseudonymous data: anyone who knows your address can compute the same checksum.
- Third country
- United Kingdom. We base the transfer on the European Commission’s adequacy decision for the United Kingdom (Article 45 GDPR), to the extent and for as long as it applies.
- Retention
- Address encrypted for at most 30 minutes, deleted immediately once redeemed. Daily counter per mailbox until two days after the day counted. Result: not stored.
Further sources for the breach check (for example the per-address search of Have I Been Pwned, XposedOrNot or Hudson Rock) are prepared but switched off. They would receive the address in plain text. Before we switch any of them on, we will update this policy.
13. Use of language models (AI)
When e-mails, messages and text messages are checked, a language model (an artificial intelligence system) looks for typical signs of fraud in the text and quotes the relevant passages. The explanations are fixed texts written by us. Where a language model is involved, we label it (Article 50 of Regulation (EU) 2024/1689).
- Data
- Subject and text of the e-mail or the text of the message, at most 12,000 characters. For e-mails we first replace your addresses and your name, as far as we can recognise them from the recipient details, and all e-mail addresses except the sender’s with placeholders. Other details in the text (for example names in the body, postal addresses, customer numbers) cannot be recognised reliably; they remain. We do not send headers, attachments, your IP address or your e-mail address.
- Purpose
- Detecting signs of fraud for the report.
- Legal basis
- Article 6(1)(b) GDPR for your data, Article 6(1)(f) GDPR for third-party data in the text (as in section 6).
- Recipients
- OpenAI (OpenAI, L.L.C., USA, or the OpenAI company responsible for customers in the EEA) through its programming interface. Some requests, mainly at night and when the daily budget is used up, are instead handled by a language model operated by Netzint at ai.netzint.de. Which service answers a given request depends on the time of day and the budget.
- Third country
- For OpenAI, the USA. We base the transfer on the EU-U.S. Data Privacy Framework, to the extent OpenAI is certified under it, and on the European Commission’s standard contractual clauses (Article 46(2)(c) GDPR).
- Retention
- With us: the model’s result appears as a quote in the report (7 days). We do not store the text sent separately. How long the provider keeps requests is governed by its terms.
Once the daily quota for the language model is reached, we analyse forwarded e-mails without a language model.
14. Data sources and lookups with third parties
We download many lists regularly and compare against them on our own systems. No data about you or your check is sent in the process. These include the Deutsche Bundesbank’s bank code file, the numbering plans and list of measures of the Bundesnetzagentur, the number ranges of the RTR, the warning lists of BaFin and AMF, the register of the European Banking Authority, Phishing.Database, the warning list of CERT Polska, lists of disposable e-mail domains and URL shorteners and the breach directory of Have I Been Pwned. The full list with operators, licences and dates is under our data sources.
For some checks a detail from the check leaves our systems:
- DNS: we look up the domains of senders and links in the Domain Name System (the domain’s name servers and those of our data centre), plus the IP address of the delivering mail server from the headers to check SPF and DKIM.
- RDAP and WHOIS: for the registration date we ask rdap.org and the responsible registry about the domain. We do not query .de domains.
- Short links: if a message contains a link from a known URL shortener, we ask the shortener where the link leads (full short link, at most three steps). We do not open the destination itself; the perpetrator’s server receives no request from us. The shortener receives our server’s IP address.
- GLEIF: if a message names a company with a legal form, we search the public LEI directory of the Global Legal Entity Identifier Foundation (Switzerland) for that name (at most two names per message).
- Warning lists queried per domain: URLhaus (abuse.ch, Switzerland) and PhishTank (Cisco Talos, USA) receive the domain of a link or sender. These lookups go through a proxy provider (Decodo, Lithuania), which also sees the domain.
- Pwned Passwords and LeakCheck: see sections 11 and 12.
None of these lookups sends your IP address, your e-mail address or the text of your message. The domains, names and links come from the checked message and concern its sender. The legal basis is Article 6(1)(f) GDPR (detecting fraud). We keep results per domain for up to six hours, GLEIF results for up to 24 hours, short link results only as domain and number of steps. Switzerland has an adequacy decision of the European Commission; for the USA, what is said in section 5 applies to the extent personal data is transferred.
15. Third-party data: senders, perpetrator characteristics and warning lists
Much of the data we process does not come from the person it concerns (Article 14 GDPR): the sender and other recipients of a submitted e-mail, phone numbers, IBANs and addresses someone has checked, and entries from authorities’ warning lists.
Characteristics of perpetrators
- Data
- If the check of an e-mail, text message or message finds phishing, fraud or malware, we store characteristics of the sender: domains, link addresses without path, IBANs, phone numbers, crypto addresses, checksums of dangerous attachments and the sender address, the latter only as a keyed value (HMAC). Also how often and when they were first and last seen, and the type of scam. No link to you as the person who submitted it is stored. We do not include recipients’ e-mail addresses.
- Purpose
- Recognising the same perpetrators in later checks and warning other users.
- Legal basis
- Article 6(1)(f) GDPR. The legitimate interest of all users and our interest is to detect fraud and prevent harm. The interest of perpetrators in not being recognised in further fraud attempts does not merit protection. If someone is recorded wrongly, for example because fraudsters misused their number or address, section 22 applies.
- Retention
- 180 days after the last sighting.
Warning lists of authorities
Under “Warnings” we show entries from lists of the Bundesnetzagentur (phone numbers against which it has taken action) and of financial supervisors (BaFin, AMF; from these lists we take only the domain of a website, no names and no e-mail addresses). We take the entries unchanged with source and date. The originator of each entry is the authority concerned. The legal basis is Article 6(1)(f) GDPR (warning about fraud). Retention: we delete AMF entries as soon as the AMF removes them from its list. The Bundesnetzagentur and BaFin publish their lists only as ongoing notices; we currently keep these entries without a fixed period, because a measure or warning still protects against fraud later on. The authorities’ own privacy notices apply to their publication.
Informing the people concerned
We do not inform senders and other recipients of a submitted e-mail or the holders of checked numbers, IBANs or addresses individually. As a rule we have no way of reaching them without collecting further data about them, and a message to suspected fraudsters would undermine the purpose of the check. We rely on Article 14(5)(b) GDPR for this and provide the information through this publicly available policy.
16. Automated assessment
Our results are produced automatically from fixed rules, lists and, where labelled, a language model. No person reviews each result. The results are an assessment without guarantee; what you do with them is your decision. We do not take any decision based solely on automated processing that produces legal effects concerning you or similarly significantly affects you (Article 22 GDPR). Liability is covered in the terms of use.
17. “Current scams” statistics
Once a day we count how many completed check reports relate to a scam type or a result. We store only the day, the scam type or result and the number, no content, senders, numbers or keys. A scam type is shown only from five checks in the period. We delete the counts after 40 days. The legal basis is Article 6(1)(f) GDPR (warning the public about current scams).
18. Contact by e-mail and operational e-mails
If you write to hello@onlinebetrug.ai, we process your address, your name if you give it, and the content of your message to deal with your request. The e-mails are held on our own mail server at Hetzner. The legal basis is Article 6(1)(b) GDPR where it concerns a check or your rights, otherwise Article 6(1)(f) GDPR (answering enquiries). We delete the e-mails once the matter is closed and no statutory retention obligation applies.
For operations our system sends notices to ourselves, for example when a data source fails. These e-mails contain no data about users and no checked content.
19. Storage on your device
We store on your device only what the feature you use needs. We do not set advertising or tracking cookies.
- Language: your language choice in the browser’s local storage (localStorage, key “lang”), until you delete it.
- E-mail and text message check: disposable address, report key and access key in session storage (sessionStorage), so the page can show you the report. The browser deletes them when you close the tab; the page discards expired addresses.
- Opening a report: if you enter your address correctly when opening an e-mail check report, our server sets a cookie (“mc_unlock”) that applies only to this report, cannot be read by scripts and expires with the report.
- CAPTCHA: Cloudflare Turnstile reads characteristics of your browser (section 5).
The legal basis for storing and reading is section 25(2) no. 2 of the German TDDDG: it is strictly necessary to provide the service you have expressly requested. We base the subsequent processing on the grounds given in the sections above. The password check and the breach check store nothing on your device.
20. Recipients and third countries at a glance
- Hetzner Online GmbH
- Hosting of all servers, database and mail servers; processor; EU.
- Cloudflare, Inc.
- CAPTCHA (Turnstile): IP address, browser characteristics; USA (section 5). Also delivers Pwned Passwords (only our server’s data).
- OpenAI
- Language model: redacted text of the checked e-mail or message; USA (section 13).
- Netzint (ai.netzint.de)
- Language model: as OpenAI, for some requests (section 13).
- LeakCheck
- Shortened checksum of your address; United Kingdom (section 12).
- Have I Been Pwned
- Five characters of a password checksum, with no link to you; Australia (section 11).
- DNS, rdap.org, registries, URL shorteners, GLEIF, URLhaus, PhishTank, Decodo
- Domains, short links and company names from the checked message; no data about you (section 14).
Beyond this we only pass on data where the law requires it, for example on the order of an authority or court (Article 6(1)(c) GDPR).
21. Your rights
You have the right of access (Article 15 GDPR), rectification (Article 16), erasure (Article 17), restriction of processing (Article 18) and data portability (Article 20). We do not ask for consent for any of the processing described; if you have given us consent, you can withdraw it at any time with effect for the future (Article 7(3)).
Right to object (Article 21 GDPR): where we process data on the basis of Article 6(1)(f) GDPR, you can object at any time on grounds relating to your particular situation. We will then stop processing the data unless we can demonstrate compelling legitimate grounds that override your interests, or the processing serves the establishment, exercise or defence of legal claims. Write to hello@onlinebetrug.ai.
We cannot link much of the data to a person: we keep no accounts, IP addresses mostly only as a keyed value that changes daily, and e-mail addresses only as keyed values. We are not obliged to collect additional data just to identify you (Article 11 GDPR). If you want information about a report or an address, give us the link to the report or the address; we can then compute the keyed value and look it up.
You can lodge a complaint with a data protection supervisory authority, in particular in the member state of your habitual residence, place of work or the place of the alleged infringement (Article 77 GDPR).
22. Corrections for people affected
If you believe that your phone number, IBAN, e-mail address or domain wrongly appears in a result or a warning, for example because fraudsters misused it, write to hello@onlinebetrug.ai. Name the detail and, if possible, give evidence. A person reviews your message. We correct or delete inaccurate entries in our own data.
If an entry comes from a list of an authority (Bundesnetzagentur, BaFin, AMF) or another source, that body is the originator of the entry. We name the source for every entry. You need to ask that body for a correction; as soon as the source changes its entry, we take this over with the next update. If an entry is obviously wrong, we can take it down until it has been clarified. For shops there is the correction page at fakeshop.ai.
23. Changes to this policy
We update this policy when we change checks, sources or service providers. Before we switch on a source that receives data about you, we update the policy. The version published here applies.
Last updated: 10 October 2026.