WHERE IS MY PHOTO
USED ONLINE?

WHY IT IS WORTH LOOKING

Searching your own face is not vanity, it is maintenance — and what turns up generally falls into three categories:

• Things you forgot. A forum avatar from ten years ago, the "our team" page of a company that closed, an event photo gallery. Harmless, but still out there without your knowledge.
• A stolen profile picture. An account using your photo but not your name. This is the most common form of fake profile and it is usually part of a scam — meaning someone is being deceived using your face.
• Commercial use. Your photo on a product page, in an advert, or as a "customer testimonial" image. Depending on jurisdiction this can breach personality rights, image rights or data protection law, and often all three.

There is a fourth and uglier category worth naming: your face placed into another image. That is no longer a technical challenge, and it is usually done for extortion. If that happens, the move is not to pay but to preserve the evidence and report it immediately.

PREPARATION: WHICH PHOTOS TO SEARCH WITH

Searching with a single frame does half the job. When looking for your own trace, use this trio:

• Your current profile picture. It is already the most widely distributed frame you have.
• A frame from five years ago. An old photo finds old accounts — and the accounts you forgot are usually the old ones.
• A passport-style frame looking straight at the camera. This is the angle face search engines handle best, and what it finds is precisely what the other two miss.

In every frame make sure the face is at least 200 pixels wide, without sunglasses or filters, and alone. Uploading a group photo without cropping directly lowers result quality.

FOUR SEARCH CHANNELS

1. Google Lens — "where was this file published?"
Upload the crop at images.google.com. What it does best is finding verbatim copies of the same image. Collect what you find in a separate tab; we will triage it shortly.

2. TinEye — "who published it first?"
tineye.com does not search faces, but it does something the others cannot: it sorts results by date. The "oldest" sort shows where an image first appeared — which is direct evidence in a copyright or unauthorised-use claim.

3. Yandex Images — "what resembles it?"
yandex.com/images has the highest similarity tolerance of the free tools. It catches cropped, filtered and resized copies more often than Google does.

4. Reverse face search — "where does this face appear?"
All three above search the file. Your actual question usually is not about the file: it is about where your face shows up. If someone is using a completely different photo of you, or edited one beyond recognition, only face search will find it. In Trace every result carries a similarity score from 70 to 100; when searching your own face, look at the 90+ band — what appears there is almost always genuinely you.

There is also Google Alerts: set an alert for your name and any handle you use. It does not search images, but if your name appears alongside your photo it will tell you about a new publication.

TRIAGE WHAT YOU FIND

Before turning the list into a removal queue, split it into three. This is the only way to spend your effort in the right place:

• Harmless and accurate. Content you posted that is still yours. Leave it.
• Unwanted but harmless. An old account, a dead page, a gallery you had forgotten. Remove it when you get to it; no rush.
• Harmful. An account impersonating you, unauthorised commercial use, private images published without consent, manipulated content. All your energy goes here.

REMOVAL: FIVE STEPS

Step 1 — collect the evidence. Link, full-page screenshot, date, and an archive snapshot if possible. Do this before reporting: once the page is taken down there is no way to produce evidence, and every subsequent step asks for these records.

Step 2 — report it to the platform. Instagram, Facebook, TikTok, X and LinkedIn all have impersonation and unauthorised image reporting forms. This is the fastest, cheapest and usually most effective route; identity verification may be requested. If the content is on a website, a report to its hosting provider often works faster than the site owner.

Step 3 — send an erasure request to the site. A photograph of your face is personal data, and where it is processed to identify you it is a special category of data under GDPR Article 9. GDPR Article 17 gives you a right to erasure, and Article 12 requires the controller to respond without undue delay and in any event within one month. Your request should include: who you are, which URL, which image, what you want (erasure) and the right you are relying on.

Step 4 — escalate if there is no response. Two routes run in parallel:

• Complain to your data protection authority. If your request is refused or ignored, your national DPA accepts complaints.
• Ask search engines to delist. Independently of the site itself, Google and Bing accept right-to-be-forgotten style requests for results about you in the EU/UK. This does not delete the page but removes it from results, which is often what actually matters.

If the photo is being used commercially, many jurisdictions also give you a separate image-rights or personality-rights claim on top of data protection law — worth naming in your request, because it raises the cost of ignoring you.

Step 5 — clear it out of search results. This step is constantly skipped, and the result is a page that has been deleted but still shows up on Google. After the page is genuinely gone, use Google's Remove Outdated Content tool to clear the cache. Image search results need to be dropped separately; they are different indexes.

OPTING OUT OF FACE INDEXES

The steps above clean up the source. One category remains: the indexes kept by face search services themselves. These services store faces they have seen on the public web on their own side, and a record can persist for a while even after the source page is deleted.

To remove that record in Trace, upload a single photo at remove my photos. Every matching entry is deleted and the face is added to a permanent do-not-index list. No account, email or ID is required, and the photo you send for the request is not stored — a removal request should not turn into a new data record.

Other services (PimEyes, FaceCheck.ID and similar) each have their own opt-out process and you have to apply to each separately; leaving one does not affect the others. We collected the comparison here.

KEEPING IT FROM HAPPENING AGAIN

• Review your profile pictures. A public profile picture can be indexed by definition. Using a high-resolution frame where your face is sharp, publicly, makes face search's job easier.
• Restrict tagging. Even if your account is private, a public post you are tagged in carries your face onto the open web.
• Search once a year. Repeat the four channels above annually. A new fake profile is far easier to remove in its first weeks than months later.
• Keep the paperwork. Archive the link, screenshot and correspondence for every piece of content you had removed. If the same content reappears, the process does not start from zero.

FAQ