What Is VeryAI? Palm Biometrics and Deepfake Detection for the AI Internet

VeryAI is the first Proof of Reality platform designed to verify whether the person on the other side of a screen is a real human. As generative AI rapidly increases the volume of bots, synthetic identities, and manipulated media online, distinguishing authentic human activity from artificial behavior is becoming a foundational challenge for the internet.

VeryAI addresses this problem by combining palm-based biometric authentication with AI and deepfake detection to create a unified trust layer for different platforms. The result is infrastructure that allows applications to verify human presence and detect manipulated content — without compromising user privacy.

The Problem

Generative AI has lowered the cost of identity fraud. Today, people can fabricate realistic identities by combining images generated by AI, cloned voices, and synthetic video that is often indistinguishable from real footage.

The security tools most platforms rely on were not designed for this environment. CAPTCHAs can be bypassed by automated systems, multi-factor authentication can be intercepted through social engineering or SIM swaps, and traditional identity checks are increasingly vulnerable to AI-assisted fraud.

Facial recognition systems face additional limitations. Human faces are widely available across social media, public datasets, and photo archives, making them relatively easy for AI models to replicate. Deepfake technology can generate highly convincing facial imagery, weakening the reliability of face-based authentication.

Research has also shown that some facial recognition algorithms produce false-positive rates up to 100 times higher for certain demographic groups, introducing fairness concerns alongside technical vulnerabilities.

As synthetic media improves and AI agents become more capable, platforms need new ways to verify real human presence — not just static identity attributes.

How VeryAI Works

Our product is a palm biometric verification system that runs on any standard smartphone camera.

The palm was selected intentionally as the biometric signal. Unlike faces or fingerprints, palm images are rarely shared publicly and are not widely stored in existing databases. Each palm contains a dense and highly unique network of lines, creases, and vein patterns, creating a biometric signature that is extremely difficult to replicate or synthesize.

During verification, the system prompts the user to perform a randomized hand gesture. This interaction confirms liveness and physical presence, preventing attacks based on screenshots, prerecorded video, or content generated by AI.

The system’s accuracy reflects the strength of this approach. A single-hand scan produces a false acceptance rate of roughly one in ten million, already exceeding the precision of many widely deployed biometric systems. When both hands are scanned, the probability of a false match drops to approximately one in one hundred trillion.

Equally important is how the data is handled. VeryAI does not store palm images. Instead, the biometric input is converted into an irreversible mathematical representation that cannot be reconstructed into the original image.

Verification produces a non-traceable identifier that confirms a real person performed an action, while preserving user privacy and preventing biometric data misuse.

Beyond Identity: Proof of Authenticity

Identity verification alone is no longer sufficient in an era of generative AI. Platforms must also determine whether content itself is authentic.

VeryAI extends its system beyond human verification by incorporating proprietary classifiers that analyze digital media for signs of manipulation. These models detect deepfake video, synthetic images, AI-generated audio, and forged documents.

When combined with palm authentication, these capabilities allow platforms to verify both who is acting and whether the content being produced is genuine. This layered approach creates a more comprehensive defense against identity fraud, impersonation, and manipulated media.

Use Cases

The ability to reliably confirm human presence unlocks new security models across multiple industries.

• Financial platforms and crypto exchanges can replace vulnerable OTP or CAPTCHA flows with palm verification, reducing bot-driven fraud and preventing synthetic account creation.
• Government agencies can use the system to ensure only verified individuals access sensitive services such as benefits, identity documents, or digital voting systems.
• Social networks and online communities can deploy human verification to reduce fake accounts, coordinated bot activity, and AI-generated impersonation.
• Healthcare portals, legal platforms, and other environments where identity certainty carries real consequences can similarly benefit from stronger authentication.

As AI-generated identities become more common, verifying that a real human is present will become a foundational requirement for secure digital systems.

VeryAI is building the infrastructure to make that possible.