Anti-Matter®
ServicesWorkAboutCareersGet in touch
All work
Creative-IP safetyProductLive2026

PowerPixel

An image-safety scanner for creators and teams: drop in a picture and PowerPixel runs five forensic signals, AI-generation detection, brand and logo presence, hidden stock-preview watermarks, perceptual fingerprinting, and content provenance, then returns one clear verdict, allowed or blocked, before you ever publish.

Visit the live site

Overview

Anyone shipping visual content now carries hidden IP risk: an image might be AI-generated, carry a brand's trademark, or still hold a stock site's invisible preview watermark. Catching that by eye is impossible, and a single bad post can mean a takedown or a legal letter. PowerPixel turns that guesswork into a fast, automated check that runs before anything goes live.

Drop an image and PowerPixel fuses five independent signals into a single result. It reads C2PA Content Credentials alongside an AI-generation classifier to flag synthetic media, matches against popular brand and trademark logos, detects hidden stock-preview watermarks, and computes a perceptual hash so near-duplicate and recycled images surface instead of slipping through. The output is one legible verdict, ALLOWED or BLOCKED, not a pile of raw scores.

Every paid scan can be sealed as an on-chain attestation: a timestamped, tamper-evident record of exactly what was checked and what the verdict was, anchored to a public ledger so a platform, counterparty, or auditor can verify it independently. The image itself never has to leave the owner's control.

Capabilities

  • Five-signal scan fused into one ALLOWED or BLOCKED verdict
  • AI-generation detection via C2PA Content Credentials and a classifier
  • Brand and trademark logo presence detection
  • Hidden stock-preview watermark detection
  • Perceptual hashing to catch near-duplicate and recycled images
  • On-chain scan attestation: timestamped, independently verifiable proof

Next project

DT-AIPTS