Age and ID verification that stores nothing.

OFCOM-ready. GDPR by design. Verify age & ID from sources your users already have : their bank, their Govt portal login. No selfies. No documents. No PII on any server.

The problem with identity verification today

Every verification creates a copy of sensitive data. Every copy is a breach waiting to happen.

Traditional (Doc + Selfie)Reclaim Protocol
User uploads government ID✅ Yes❌ No
Selfie / biometric collected✅ Yes❌ No
PII stored on server✅ Yes. Retained up to 3 years❌ Never touches a server
Breach risk🔴 High. Data held on server🟢 Zero. There's nothing to leak
Spoofable by minors🔴 Yes. Screen scanning, AI-generated docs🟢 No. Proof comes from authenticated session
Requires app downloadSometimes❌ No. Works fully on browser
GDPR data minimization⚠️ Difficult. You typically hold more data than you need✅ By design. Only the absolutely required data is extracted

How it works: 30 seconds. 3 steps.

1

User logs in

The user logs into a trusted source they already have an account with: a bank, mobile carrier, health portal, or government service. Reclaim never sees their credentials.

2

The data is extracted

Only the specific data point you need is selected — e.g., "age = 42". Nothing else is extracted. Nothing else is shared.

3

A cryptographic proof is generated

A zero-knowledge proof is created on the client's device, verifying the data came from the claimed source and hasn't been altered. The proof is cryptographically tied to the source's TLS certificate. Tamper-proof. Unforgeable.

How we compare

YotiPersonaReclaim Protocol
MethodSelfie age estimationDocument upload + selfieLogin into govt portal, bank etc
PII collectedFacial biometricsGovt ID + faceNone
Data storedYes — processed on serverYes — claims 3-day deletionNothing. Ever.
Deepfake-proofNoSelfie estimation can be spoofed with AI-generated facesNoDocuments and selfies can be forged with AIYesProof is from an authenticated TLS session. Nothing to forge.
Known incidents€950K fineSpain's AEPD fined Yoti for biometric data handling, invalid consent, and excessive data retention70K IDs exposed~70,000 government-issued ID images exposed via customer service vendor. Discord publicly cut ties with Persona.NoneNo personal data is collected, processed, or stored. Nothing exists to breach.
If breached, what leaks?User face scansGovernment IDs, selfiesNothing. No data exists.
Your liability if vendor is breachedYou're liableYou're the data controller. Biometric data is GDPR special category.You're liableYou're the data controller. ID documents are personal data.No liabilityNo personal data was processed. No breach notification required.
GDPR burdenHighBiometric data = special category. Requires explicit consent + DPIA.HighID document retention requires lawful basis + data minimization proof.MinimalNo personal data processed. Data minimization by design.
Integration timeDays to weeksDays to weeksHours

Read our blog on how to comply with OFCOM and GDPR in tandem

Try it out yourself

Countries being added to coverage at a rapid pace. Learn More

Regulatory FAQ

CountryRegulationStatusNote
🇬🇧 United KingdomOnline Safety Act (OFCOM)Aligns with "reusable digital identity"
🇪🇺 European UnionDigital Services Act (Art. 28)Matches EU zero-disclosure blueprint
🇫🇷 FranceLoi SREN (Arcom)Fits double-anonymity principle
🇩🇪 GermanyYouth Protection Act (KJM)⚠️Requires formal KJM assessment
🇮🇳 IndiaDPDP Act & Rules 2025DigiLocker/Aadhaar explicitly approved
🇦🇺 AustraliaOnline Safety Act (eSafety)☑️Awaiting eSafety technical standards
🇧🇷 BrazilDigital ECA (Lei 15.211)☑️Govt ID explicitly permitted
🇺🇸 United StatesCOPPA (FTC)☑️FTC incentivizes age verification
🇨🇦 CanadaBill S-210 / Bill C-63Privacy-preserving explicitly required

Verify identity without fraud and data breaches

Verify identity, age, and citizenship directly from government portals. No documents to review, no fraud to worry about.