Protect your family from deepfakes, scams, and impersonation.
Enroll the people you are authorised to protect, monitor potential likeness misuse through an approved provider, review evidence, and prepare action from one private workspace.
Start free · No card required · You control consent and deletion
Consent recorded
Before biometric enrollment
Private gallery
Owner-scoped access
Signed discovery
Verified provider events
Action trail
Audited reviews and reports
Comprehensive likeness protection
Monitoring workflows, consent controls, evidence review, and a complete image privacy toolkit.
Continuous likeness monitoring
Consent-first enrollment and provider-backed discovery workflows for finding potential online uses of a protected likeness.
Matches, alerts, and action
Review candidate matches with confidence and threat signals, mark authorised use, and prepare tracked abuse reports.
Family protection
Protect up to five household members with explicit consent state and guardian attestation for minors.
Metadata remover
Strip EXIF, GPS, camera, and hidden metadata in bulk — with a full field-by-field breakdown of exactly what was removed.
Privacy scanner
Detect faces, license plates, QR codes, barcodes, documents, and text regions, with a clear risk score and guidance.
One-click redaction
Blur, pixelate, or black-box detected faces, plates, and text — automatically or with manual control over style and strength.
Protection pipeline
A modular pipeline — metadata removal, normalization, subtle transforms, watermark, compression — each stage you control.
File fingerprint
Generate a fresh file identity — new name, size, type, and hash — while keeping the picture intact.
Location cleanup
See GPS coordinates on a map and remove them, so your photos stop revealing where they were taken.
Invisible watermark disruption
Degrade SynthID-style watermarks hidden in the pixels themselves — measured with SSIM and a perceptual-hash shift, and reported honestly as disruption rather than guaranteed removal.
AI-provenance report
Three separate signals, kept separate: embedded markers (C2PA, IPTC, generator tags), pixel-level classification, and the watermark itself. No single misleading "AI: yes/no".
Visible watermark removal
Select the area covering a stamped logo or caption and reconstruct it from the surrounding pixels, with a similarity score so you can see what it cost.
Adversarial face cloaking
Optional perturbations aimed at face-recognition models, verified against ArcFace on the saved file — the cosine similarity is measured and shown, not asserted.
Batch processing
Upload and process many images at once, then export cleaned copies as files or a ZIP with your own naming pattern.
Prepaid image tools
Buy credits for image operations and keep what is left. Credits never expire; likeness monitoring is a separate, clearly disclosed plan.
How it works
Three clear steps from consent to action.
Enroll your household
Add yourself or an authorised family member and record the required biometric consent.
Monitor
An approved provider searches configured sources and returns candidate likeness matches.
Review and act
Inspect evidence, classify authorised use, and prepare a platform abuse report when action is needed.
Scan, then actually fix it.
Most tools only tell you what's wrong. ImageShield AI lets you act: every finding comes with one-click fixes — blur faces, mask plates, redact text, remove GPS — plus manual control when you want it. Re-scan to watch the risk score drop.
- Field-by-field metadata removal with before/after comparison
- Automatic or manual redaction (blur, pixelate, black-box)
- Configurable protection pipeline you control stage by stage
- Invisible-watermark disruption with measured SSIM and hash shift
- AI-provenance report across markers, pixels and watermark
- Downloadable PDF & JSON privacy reports
Privacy by construction
Your face is sensitive data. We treat it that way.
Every protected person has explicit consent state, every reference image stays tied to its owner, and withdrawal retires active biometric enrollments.
Read the privacy policyAccess controlled
Authentication, tenant isolation, role controls, rate limits, and audited security events.
Deletion controls
Withdraw consent or remove a person and active enrollment records are retired immediately.
Purpose limited
Reference photos are used for your protection workflow, never advertising or general identity search.
Evidence, not guesses
Confidence and deepfake signals remain separate and require a human review before action.
Image tools with prepaid credits.
Buy a pack of credits, spend them on the images you actually process, and keep whatever is left. Credits never expire and do not renew automatically.
What a credit buys
- Remove metadata1 credit
- Privacy scan1 credit
- AI-detection check1 credit
- Disrupt invisible watermark1 credit
- Full protection pipeline2 credits
- Adversarial face cloaking3 credits
A typical photo — cleaned, scanned and protected — costs 4 credits.
Packs from $10
$10 covers about 60 images end to end, and the price per credit drops as the packs get larger — down to 2.5¢ on the largest. Every pack unlocks every tool; the only difference is how much you can do.
New accounts start with 20 free credits, waiting on the dashboard and unlocked by your first top-up.
Secure by default
JWT auth with refresh rotation, account lockout, optional two-factor (TOTP), audit logging, rate limiting, and CSP. Your files are processed for you, never sold.
Honest about privacy
We never claim an image becomes impossible to identify or that we defeat a specific recognition service. Every feature describes a realistic improvement — and shows you exactly what changed.
Ready to protect your digital image?
Create a free account, enroll your household with consent, and start improving image privacy in minutes.