Of every photo of your face that lives on the public internet, the one you posted to LinkedIn is the one a face-recognition model would most want.
- Frontal pose. Eyes level, shoulders square. The angle the model was trained on.
- Even, soft light. No deep shadows hiding cheekbones or jaw geometry.
- Slight smile. Mouth slightly open, no shoes, no sunglasses, no hat. Every landmark visible.
- High resolution. LinkedIn supports up to 8 MB upload. Most people upload 1024+ on the long edge.
- Same face every photo update. You change headshots every few years — perfect for tracking aging.
- Public by default. LinkedIn profiles are indexed by Google. Profile photos are crawlable without login.
If you were designing the perfect training image for a face-recognition system, you would design exactly this. And about a billion people have already uploaded it for you.
Who is scraping LinkedIn for face data?
More or less everybody. Some of them are open about it; most aren't. The short list:
Face-search engines (PimEyes, FaceCheck.ID, Lenso.ai)
LinkedIn profile photos appear constantly in their search results. They scrape what's publicly viewable — and LinkedIn's profile photos are publicly viewable without login.
Clearview AI
Clearview's claimed ~50B-image index pulls heavily from LinkedIn. The company has been sued by LinkedIn's parent Microsoft over the scraping, lost in some jurisdictions, kept scraping anyway.
Candidate-screening vendors
A growing tier of HR vendors run face match against candidates' LinkedIn photos as part of "identity verification" or "fraud prevention" during interviews. Most candidates have no idea their photo is being matched against anything other than their own resume.
OSINT investigators (and stalkers)
The first thing anyone with a LinkedIn-locked photo does is paste it into PimEyes or FaceCheck.ID. It works because the LinkedIn photo is the clean baseline that matches everything else online.
The HR-tech part nobody mentions
In 2025 there was a quiet shift in HR-tech. Vendors that used to do background checks started adding "face match" as a default. The pitch to employers: "verify that the candidate on Zoom is the same person as the resume."
In practice, the workflow looks like this:
- Candidate accepts an interview.
- HR vendor pulls the candidate's LinkedIn photo as a "reference."
- During the Zoom interview, the vendor's tool captures a frame of the candidate.
- Face match runs. If the match is below a threshold, the candidate gets flagged — sometimes silently — for "verification issues."
Candidates almost never get told this is happening. The vendor's privacy policy usually buries it under "biometric data processing for fraud prevention." The HR team that hired the vendor often doesn't fully understand what the tool does either.
So just take down your LinkedIn photo?
You could. It would help going forward. It would not help retroactively, because:
- Cached copies exist. The Internet Archive, Google's image cache, and every face-search engine's local copy still hold the version that was up when they crawled.
- Your profile is the index key, not just the photo. Engines associate the photo with the name + employer + location on the profile. Even after you take down the photo, that linkage is in their database.
- Recruiters expect a photo. Removing it changes your professional profile's signal in ways most people don't actually want.
- Other photos of you exist. Conference panel shots, company "About" pages, news mentions. The LinkedIn photo is the most-scraped but it's not the only one.
Removing your LinkedIn photo is a small win. The bigger win is removing yourself from the engines that already indexed it.
What we'd actually do
Honest answer: keep the LinkedIn photo (recruiters expect it; the professional cost of removing it is real), and file removals against the engines that are indexing it.
The engines respond to removal requests under the privacy law that covers you (GDPR, CCPA, UK DPA, etc.). They'll re-scrape next month and pull your face back in if you're still publicly visible. So removal has to be ongoing, not one-shot.
File the removals yourself once a month — or hand the whole list to us and the refile cadence comes built-in.
What about the HR-tech vendors specifically?
Harder to opt out of, because you usually don't know which one a given employer uses. Three things help:
- If you live somewhere with biometric-consent laws (Illinois BIPA, Texas CUBI, Washington biometric, EU GDPR Art. 9), employers using these tools without your explicit consent are in legal jeopardy. More on jurisdictional rights here.
- Ask before the interview. "Does this interview involve any biometric verification or facial recognition tools?" Most HR teams won't lie if asked directly.
- Remove yourself from the underlying indexes. Most HR vendors don't have their own face database — they query someone else's. If you're not in the queried index, the match fails because there's nothing to match against.
The headshot is the gift you keep giving — to recruiters, and to everyone else.
You don't have to take it down. You do have to make sure the engines that scraped it stop returning you when someone searches.
Start Removal →Two-week free trial. Cancel anytime. ♡