Slashdot Reader Builds a Free Photo-Verification App for iPhones

Veridenz introduces a workflow where a photo is locally hashed, the resulting digest is signed using a key that the app creates and stores in the iOS Keychain, and the signed digest is broadcast to Nostr relays. This separates the proof from the image data, ensuring that the raw file never leaves the device unless the user explicitly shares it. The reliance on the Keychain means the private signing material benefits from the operating system’s secure enclave. The published witness is a small, immutable record that can be retrieved later for comparison.

For developers, the app provides a ready-made mechanism to attach a verifiable identifier to any image without building custom cryptographic pipelines. Integration points include fetching the signed hash from Nostr relays and comparing it against a newly computed hash of a candidate file. Because the signing identity is generated by the app, there is no need to manage external credentials, simplifying onboarding for end users. However, any system that consumes these proofs must be able to parse Nostr messages and verify signatures against the public key derived from the app’s identity.

Adopting Veridenz incurs minimal direct cost: the app is free and runs on any iPhone, and no subscription or Nostr account is required. The primary operational expense is the developer time needed to incorporate Nostr relay queries and signature verification into existing workflows. Engineers must also consider key lifecycle management, such as handling key rotation or revocation if the device is compromised, because the trust model hinges on the secrecy of the stored private key.

The solution’s scope is limited to confirming that a file matches the originally witnessed hash; it does not provide any guarantee about the truthfulness of the captured scene. Consequently, it cannot replace forensic analysis or contextual verification. Additionally, the proof depends on the availability of Nostr relays; if those nodes are unreachable, new witnesses cannot be published, and existing ones may become harder to retrieve. Finally, the approach is confined to iOS devices, so cross-platform workflows will need alternative mechanisms for generating comparable proofs.