OPEN_SOURCE ↗
REDDIT · REDDIT// 27d agoPRODUCT LAUNCH
Brevis launches Vera for cryptographic media provenance
Vera is a media authenticity system from Brevis that combines C2PA hardware signatures with zero-knowledge proofs to mathematically prove where an image or video originated and what edits were made. Unlike probabilistic deepfake detectors, it shifts the burden of proof — authentic media proves its own origin, no AI classifier needed.
// ANALYSIS
The deepfake detection arms race has a structural flaw: every AI detector is one model update behind the attackers. Vera's ZK-proof architecture may be the first credible escape from that loop.
- –Builds on C2PA — the hardware-level media signing standard already backed by Adobe, BBC, Intel, Sony, and Nikon — then adds a zero-knowledge proof layer for the post-capture editing workflow
- –ZK proofs let editors demonstrate they applied only permitted transformations without exposing raw content or workflow details, giving publishers a privacy-preserving chain of custody
- –Open-source reference implementation lowers the barrier for newsrooms and platforms to integrate without vendor lock-in
- –The real constraint is hardware: C2PA support in cameras and phones is still sparse, so most real-world media can't start the trust chain yet
- –Competes structurally with Adobe Content Credentials and Truepic but claims a unique ZK-over-C2PA stack that neither rival has matched
// TAGS
brevis-veraopen-sourcesafetyethicsresearch
DISCOVERED
27d ago
2026-03-15
PUBLISHED
27d ago
2026-03-15
RELEVANCE
7/ 10
AUTHOR
renkure