Multi-factor authentication for people. Verify who's real before every call, message, or video chat.

Deepfakes are everywhere.

Voice clones. Video fakes. AI-powered impersonation scams.

Every day, families lose millions to scammers using AI to clone voices and faces. Companies wire money to fake executives. Parents panic over kidnapping hoaxes. Grandparents send life savings to fraudsters pretending to be their grandchildren.

How do you know who's real anymore?

Traditional authentication—passwords, 2FA, even biometrics—can all be faked, phished, or deepfaked. When a "CEO" video calls asking for an urgent wire transfer, or "your daughter" calls crying about an emergency, how do you verify it's really them?

The internet never learned how to tell who's real. We did.

realxreal is multi-factor authentication for people.

realxreal is a trusted contacts app where each connection is verified through shared memories you've lived together. Instead of passwords, phone numbers, or server checks, your phone recognizes people the same way you do—through shared experiences only real people know.

Before a call, message, or video chat begins, the app confirms that both people are who they say they are—instantly, securely, and privately.

Every call. Every message. Every video chat.

Who's really on the other end?

realxreal transforms shared human memories into unforgeable cryptographic keys. When someone tries to contact you:

1. Your phone sends a challenge based on a memory you've shared
2. Only the real person can answer correctly
3. Authentication happens in seconds, on-device, with zero server involvement
4. You know it's really them—not an AI, not an impersonator

Real Recognizes Real AI exists to establish & maintain trust among people, organizations, and entities that are vulnerable to identity-based attacks by empowering them to authenticate all of their digital contacts & most trusted real-life relationships.

A world where real people can confidently communicate with real people without fear of fraud, impersonation, or manipulation.

Our core technology is The Doppelgänger Protocol™ is a cryptographic approach that transforms shared human memories into unforgeable authentication keys.

Accepted at OSCW 2026 - Our research talk, "Breaking Deepfakes With Shared Secrets: Building The Doppelgänger Protocol," was accepted at the Open Source Cryptography Workshop in Taipei.

• Memory-Based Authentication - Shared experiences become cryptographic proof of relationship
• Local-First Architecture - All verification happens on-device, no server dependency
• Zero-Knowledge Principles - Privacy-preserving challenge-response protocols
• AI-Resistant - Fundamentally unfakeable by deepfakes or synthetic identity
• Human-Centric Design - Works the way real people verify identity naturally

Challenge Generation → Memory Query → Cryptographic Proof → Identity Verified
         ↓                   ↓                  ↓                    ↓
   On-Device AI      Real Human Only    Local Verification    Zero Trust

The protocol combines:

• Challenge-response cryptography - Dynamic, unpredictable verification
• Local key generation - No centralized authority or trust assumption
• Privacy preservation - Memories stay private, only proof is shared
• Multi-modal verification - Voice, video, text—all authenticated

• Grandparent scams - Verify it's really your grandchild before sending money
• Emergency calls - Confirm family members in crisis situations
• Daily communication - Peace of mind that calls and messages are authentic

• Executive impersonation - Stop BEC (Business Email Compromise) attacks
• Wire transfer verification - Authenticate high-value transaction requests
• Remote team security - Verify colleagues on video calls and messages

• Account recovery - Prove identity without phishable security questions
• High-risk approvals - Multi-person authentication for critical decisions
• Emergency protocols - Verify identity when primary systems are compromised

We believe identity protection should never depend on blind trust.

That's why our core technology, The Doppelgänger Protocol, will be open source and publicly auditable. Security through obscurity is a lie. True security comes from transparent, verifiable systems that the community can examine, audit, and improve.

• Transparent Methods - Every cryptographic decision is documented and auditable
• Community-Driven - We welcome contributions, critiques, and improvements
• No Vendor Lock-In - Anyone can implement, fork, or extend the protocol
• Public Good - Identity protection should be accessible to everyone
• Academic Rigor - Research-grade cryptography, peer-reviewed and published

Security and transparency are not opposites—they are inseparable.

Every design decision, every protocol, every line of code follows these principles. Trust can't be demanded, it must be earned and verified.

We build systems that prove real human relationships, not synthetic ones. Our technology distinguishes between people and AI because human connection matters.

Your memories stay yours. We use zero-knowledge cryptography to verify identity without exposing personal information. No servers, no surveillance, no data harvesting.

Identity protection shouldn't require technical expertise. We make cryptographic security accessible to grandparents, small businesses, and everyone in between.

We're building for the long term—technology that works 10 years from now, not just today. Our open source approach ensures longevity beyond any single company.

Cryptographic rigor in every layer. We don't compromise on math, and we don't invent our own crypto. We use proven primitives in novel combinations.

We serve people vulnerable to identity attacks—families, businesses, organizations. Their safety drives our decisions, not growth metrics or investor pressure.

Open source core. Published research. Public roadmap. We show our work because secrecy breeds distrust.

Current Phase: Active Development & Research Validation

• ✅ Doppelgänger Protocol research accepted at OSCW 2026
• ✅ Prototype application demonstrating core verification flow
• ✅ Initial cryptographic architecture documented
• ✅ Privacy-preserving challenge generation system
• 🚧 Mobile-first reference implementation
• 🚧 Multi-device synchronization protocol
• 🚧 Open source protocol specification

• OSCW 2026 presentation (March 2026, Taipei)
• Open source protocol release
• Reference implementation for developers
• Community beta testing program

• "Breaking Deepfakes With Shared Secrets: Building The Doppelgänger Protocol"
  Open Source Cryptography Workshop (OSCW) 2026, Taipei
  Jonathan Bateman

• Technical specification of The Doppelgänger Protocol
• Privacy analysis and threat modeling
• Usability studies on memory-based authentication
• Comparative analysis vs. traditional MFA

We welcome contributions from cryptographers, developers, security researchers, and domain experts. Areas where we need help:

• Cryptographic Review - Audit and improve our protocol design
• Privacy Engineering - Strengthen zero-knowledge implementation
• Usability Research - Make cryptography accessible to non-technical users
• Threat Modeling - Identify and address edge cases
• Documentation - Help others understand and implement the protocol

Coming Soon: Contribution guidelines and security disclosure process

The Doppelgänger Protocol and reference implementations will be released under an Open Source Initiative (OSI) approved license to ensure auditability, transparency, and community ownership.

Different components serve different purposes:

• Protocol Specification: Released under permissive license (MIT/Apache 2.0) for maximum adoption
• Reference Implementation: Open source core with commercial-friendly licensing
• Mobile Applications: Source-available for audit, with appropriate app store compliance
• Research Papers: Published under open-access terms (Creative Commons)

Note: As we transition toward public release in 2026, specific license files will be added to each repository to provide definitive legal clarity.

• ✅ AI voice cloning and deepfake video impersonation
• ✅ Synthetic identity and account takeover attacks
• ✅ Social engineering leveraging stolen personal information
• ✅ Compromised credentials (passwords, 2FA, biometrics)
• ✅ Real-time deepfake attacks during live communication

• ❌ Coerced authentication (someone forced to answer under duress)
• ❌ Memory degradation over time (system includes refresh mechanisms)
• ❌ Completely fabricated relationships (no shared memories exist)
• ❌ Device compromise (assumes trusted device environment)

• Users can distinguish real memories from fabricated ones
• Devices are reasonably secured (standard mobile OS protections)
• Cryptographic primitives (AES, elliptic curves) remain secure
• Adversaries cannot extract memories through social engineering at scale

• $2.8 billion lost to BEC and executive impersonation in 2023 alone
• 35% increase in deepfake fraud attempts year-over-year
• Millions of families targeted by grandparent scams and emergency fraud
• Every major organization now faces AI-powered impersonation risk

Traditional authentication can't stop this. Passwords get phished. 2FA gets bypassed. Biometrics get deepfaked.

We need a new layer of trust built on something AI fundamentally cannot replicate: shared human experience.

• Review our OSCW paper and provide feedback
• Collaborate on protocol extensions and improvements
• Help us publish rigorous security analysis

• Implement the protocol in your applications
• Build integrations and tools
• Report vulnerabilities responsibly

• Pilot the technology in high-risk authentication scenarios
• Share real-world threat intelligence
• Help us understand your security needs

• Star this repository to follow our progress
• Share our mission with people vulnerable to identity fraud
• Engage in discussions about cryptographic identity verification
• Spread awareness about the deepfake threat

• Website: realxreal.ai
• GitHub: github.com/realxreal-ai
• Research: OSCW 2026 paper (coming soon)

• Open source cryptographic protocol for human authentication
• Privacy-preserving identity verification system
• Research-backed defense against AI impersonation
• Community-driven security infrastructure

• Surveillance technology or mass monitoring system
• Blockchain or cryptocurrency project
• Replacement for all authentication methods
• Magic solution that requires zero user participation

We are committed to ensuring our technology is used ethically:

• Protects individuals and organizations from fraud
• Preserves privacy through zero-knowledge cryptography
• Operates transparently with open source code
• Serves the public good, not surveillance or control

Misuse Notice: This technology should not be used for mass surveillance, coerced authentication, or any application that violates human rights or civil liberties.

The internet was built without a way to prove who's real. We're fixing that.

Not with biometrics that can be faked.
Not with passwords that can be stolen.
Not with centralized authorities that can be compromised.

With shared human experiences that AI fundamentally cannot replicate.

Real people recognizing real people, the way we always have—through memories we've lived together.