Defend your Human essence
The first biometric firewall for the generative era. Secure your likeness, control your digital signature, automate unauthorized media takedowns, and manage compliance, consent registry, and AI usage licensing — all from one identity vault.
The end of trust in visual evidence.
Faces, voices, and personalities can now be cloned from seconds of public footage. As generative models reach parity with reality, the burden of proof shifts. Without a cryptographic anchor, your likeness is vulnerable to weaponized synthesis.
By 2026, the majority of media online will be AI-touched. Manual verification is no longer possible at scale.
Voice clones bypass bank security. Face clones bypass KYC. Video clones move markets and elections.
Most victims discover misuse weeks after damage is done — with no infrastructure to fight back.
Six modules. One trust layer for the AI internet.
Vault. Firewall. Detection. Each layer reinforces the others — every signal you generate hardens the system that protects you.
Identity Vault
Encrypted storage of your biometric ground truth — faceprints, voiceprints, signatures. Your reference data never leaves the enclave.
Consent Engine
Define exactly how your likeness can be used. Allow parody. Block NSFW. Block political. License commercial usage on your terms.
AI Firewall (UAIP)
A real-time API that AI providers call before generating media of you. Returns allow / deny with a cryptographic verdict.
Deepfake Scanner
Reverse search across social platforms and the open web. Detect AI generations of your face, voice, and brand within minutes.
Authenticity Verification
C2PA-style provenance signatures and invisible watermarking. Prove your real content is genuinely yours.
Takedown Toolkit
Auto-generated DMCA notices, platform reports, and court-ready evidence packs. Reclaim your identity in days, not months.
Four steps from anonymous target to protected identity.
- 01
Register
Upload reference face, voice, and brand assets to your vault.
- 02
Set rules
Choose what AI systems may and may not generate of you.
- 03
Providers check
Partner AI models call UAIP before generation. Unauthorized requests are denied.
- 04
We detect & act
Continuous scans flag misuse. We file takedowns. You stay protected.
Beyond protection. Governance, licensing, and compliance.
Consent Registry
A living, auditable record of how your likeness may be used. Set rules, grant licenses, and revoke access at any time.
AI Usage Licensing
Monetize your identity safely. License your likeness to approved creators, brands, or AI models with cryptographically signed terms.
Compliance & Audit
SOC 2, GDPR, and BSI-ready audit trails. Every vault access, consent change, and takedown is logged and exportable.
Legal Evidence Pack
Court-admissible bundles of hashes, timestamps, detection reports, and platform correspondence — ready for your counsel.
Identity Arbitration
Dispute resolution between you and AI providers. Our neutral registry mediates conflicts over unauthorized generation.
Enterprise SLAs
Custom contracts, on-premise vaults, dedicated support, and mission-critical uptime guarantees for governments and banks.
Brand Protection API
Bulk monitoring and takedown for corporate brands, celebrity portfolios, and talent agencies.
Training Data Opt-Out
Register your biometric fingerprint with partner model trainers. Prevent your likeness from being used in future model training.
The UAIP Protocol
Universal AI Identity Protection. A low-latency gateway any generative pipeline can call before producing media of a human. Open, free to implement, signed end-to-end.
// Before generating, ask the firewall
const verdict = await th.check({
actor_id: 'model:gen-ai-01',
biometric_hash: 'sha256:8f3c…',
purpose: 'commercial-ad'
});
// Response
{ status: "ALLOW", signature: "verified_human:0x7f…" }
For creators & public figures
Automate detection and removal of deepfakes across 500+ platforms. Register your official likeness so only approved content can be generated.
- 01Auto-DMCA for synthetic voice clones
- 02Cryptographic watermarking for raw footage
- 03Verified-human badge for your authentic posts
Honest pricing. 7-day free trial on every plan.
Personal protection. For everyone.
- + 1 identity
- + Monthly scan
- + 3 takedowns / month
- + Email alerts
For creators and public figures.
- + 3 identities
- + Weekly monitoring
- + 25 takedowns / month
- + Verified-human badge
Agencies, brands, and executive protection.
- + 25 identities
- + Daily monitoring
- + Unlimited takedowns
- + Team seats + API
For governments, banks, and media.
- + Unlimited identities
- + SSO + audit logs
- + Dedicated counsel & SLAs
- + On-prem options
Questions, answered.
How accurate is your deepfake detection?+
We combine multiple detection providers, biometric matching against your registered vault, and provenance signal analysis. No single detector is perfect — but layered scoring is dramatically more reliable than any one tool alone.
What data do you store, and where?+
Only the embeddings (mathematical fingerprints) of your reference media — never the original files in a usable form. Reference assets are encrypted at rest and never shared with third parties.
Which AI providers support UAIP today?+
We are in active conversations with major model providers. The protocol is open and free to implement. Detection works against any model, including ones not yet integrated.
How fast are takedowns?+
Median time to first platform action is under 72 hours for tier-1 platforms. We auto-generate evidence packs to accelerate the process.
Can I be falsely flagged?+
Deep Ban does not block real content. The Firewall API only blocks generation that matches your registered identity hash and violates the rules you set.
The future is verifiable.
Join the infrastructure layer for the next era of media integrity. Limited waitlist openings.
Join the protocol waitlist