Facial recognition technology has moved from science fiction to everyday surveillance infrastructure. Law enforcement agencies, retailers, and private companies now routinely use it to identify people from photos and video — often without consent. Understanding how your face enters these databases and how to limit your exposure is an increasingly important privacy skill.

How Your Face Enters Databases

Companies like Clearview AI scrape billions of photos from Facebook, Instagram, LinkedIn, Twitter, and other public platforms. These photos train facial recognition models that can then match your face from any photo — including CCTV footage.

Clearview AI’s database reportedly contains 30+ billion faces scraped from the internet.

Public Photos and Tagging

Every tagged photo on Facebook or Instagram trains Meta’s facial recognition system. Even without being tagged yourself, appearing in a friend’s public photos contributes to your facial profile in scraped databases.

Government and Law Enforcement

DMV photos in the US are frequently shared with law enforcement facial recognition systems. Passport photos, TSA PreCheck enrollment, and Global Entry all create government biometric records.

Opting Out of Commercial Facial Recognition

Clearview AI Opt-Out

Clearview AI is legally required to honor opt-out requests in some jurisdictions:

Visit clearview.ai/privacy
Submit a Privacy Rights Request
Select your applicable rights (CCPA if California resident, GDPR if EU resident, general privacy request elsewhere)
Submit a selfie-style photo and your email — they use it to locate and suppress your facial data

Suppression means your face is flagged to not appear in search results, but your data may still be retained.

PimEyes Opt-Out

PimEyes is a public facial recognition search engine. Opt out:

Visit pimeyes.com → Privacy
Submit a removal request with your photo
EU residents can file GDPR erasure requests

Meta Facial Recognition

Meta disabled automatic facial recognition in 2021 and deleted its database. However, facial recognition may still be used for other features. Review your Privacy settings → Face Recognition if available in your region.

Remove Your Photos from Data Broker Sites

Facial recognition is only as effective as its source photos. Removing your photos from data broker sites limits the training data:

Manual opt-outs:

Spokeo.com/optout
Whitepages.com (search your name → Opt Out)
Intelius.com (search → Remove My Info)
BeenVerified.com/opt-out
PeopleFinder.com (search → Opt Out)

Automated removal services:

DeleteMe ($129/year): Monitors and removes from 30+ sites
Privacy Bee ($197/year): More comprehensive, 200+ sites
Kanary ($9/month): Ongoing monitoring

Remove EXIF Data Before Posting

As covered in our metadata guide, strip GPS and device data from photos before sharing online. While this doesn’t prevent facial recognition, it prevents location data from being associated with your face.

exiftool -all= -overwrite_original photo.jpg

Reduce Tagging and Indexed Photos

Set social media profiles to Friends only or fully private
Untag yourself from others’ photos where possible
Ask friends not to tag you in public posts
Regularly audit tagged photos on Facebook: Activity Log → Photos → Tagged Photos

Limit High-Quality Face Photos

Facial recognition accuracy increases with high-quality, front-facing, well-lit photos. Avoid posting professional headshots to public profiles if privacy is important. Use lower-resolution, angled, or partially obscured photos for public-facing profiles.

Physical Countermeasures

Infrared LEDs

Cameras using infrared sensors (common in CCTV) can be blinded by infrared LEDs worn in glasses frames or hats. IR LED privacy glasses create a glare invisible to the human eye but overexposing the camera sensor where your eyes appear. Effectiveness varies significantly by camera type and distance.

CV Dazzle Makeup

Academic research by Adam Harvey demonstrated that specific geometric makeup patterns can disrupt facial landmark detection algorithms. While impractical for daily use, it highlights that facial recognition isn’t infallible.

Hat and Sunglasses

The most practical physical countermeasure: wide-brimmed hats combined with sunglasses significantly reduce recognition accuracy in top-down mounted cameras (common in retail). Less effective against direct-facing cameras.

Legal Protections (US)

Several states have biometric privacy laws:

State	Law	Key Rights
Illinois	BIPA (Biometric Information Privacy Act)	Strongest — requires consent before collection
Texas	CUBI	Requires consent for biometric capture
Washington	My Health MY Data Act	Health context biometrics
California	CCPA/CPRA	Right to opt out of sale of biometric data

If you’re in Illinois and a company collected your biometric data without consent, you may have a BIPA claim.

Under GDPR Article 9, biometric data used to uniquely identify people is a special category of data requiring explicit consent or another specific legal basis. EU citizens can:

Request deletion under Article 17 (Right to Erasure)
File complaints with their national Data Protection Authority (DPA)
Sue companies in member state courts

Clearview AI has been fined and ordered to delete data by EU and UK regulators.

Realistic Expectations

Complete facial recognition opt-out is impossible — your face appears in countless contexts you can’t control: store surveillance, CCTV, photos taken by others in public spaces, government systems.

The realistic goal is reducing your profile in commercial databases and preventing easy identification from your voluntary public photo presence. Removing yourself from data broker sites and Clearview AI is worthwhile and actionable. Wearing a hat in public is reasonable. Refusing to appear in photos entirely is impractical for most people.

Focus privacy efforts proportionally to your actual threat model — most people’s primary concern should be data brokers and targeted advertising, not government facial recognition systems.