No proof of origin
A file's metadata can be stripped or forged in seconds, so "I made this first" is usually unprovable once content leaves your phone.
A blockchain-based artificial intelligence system designed to help identify, verify and respond to the misuse of digital media. Hikmaon anchors your photos, videos and audio on a hybrid blockchain, issues a signed certificate of ownership, then looks for edited copies and deepfakes of your content.
Built for early users, researchers and institutional pilot partners while the platform is validated.
Full project title Hikmaon: a blockchain-anchored system and method for digital media authenticity verification, deepfake detection, ownership proof, and automated rights-based Notification
The problem
Generating a convincing fake face, voice or video takes minutes and no expertise. The damage is not only that fakes exist; it is that the real owner has no practical way to prove what they actually created, or to find copies once they spread.
A file's metadata can be stripped or forged in seconds, so "I made this first" is usually unprovable once content leaves your phone.
Re-encoded, cropped and filtered reposts no longer match the original byte for byte, so ordinary search finds nothing.
By the time a person finds a deepfake of themselves and files a notice by hand, it has usually been copied again.
How Hikmaon works
You connect the places your content already lives. Hikmaon fingerprints it, anchors the proof, watches for copies, and acts only on your decision.
Link Instagram, Facebook, Snapchat, X, YouTube, TikTok, LinkedIn, Reddit, Google Drive, Dropbox or OneDrive through standard OAuth. Nothing is uploaded without you connecting it.
Each item gets a SHA-256 hash, perceptual fingerprints and an AI embedding. The proof is anchored on Hikmalayer and you receive an Ed25519-signed Certificate of Ownership that anyone can verify.
An autonomous, robots.txt-respecting crawler scans public pages, social posts, blogs and forums. Suspected copies are scored as a 0–100% match against your anchored originals, with manipulation forensics on top.
You are alerted and you decide: Allow leaves it up; Remove files a DMCA-style takedown case automatically, with the blockchain evidence attached and tracked to resolution.
Use cases
The same anchor-and-verify core works on any file whose authenticity matters. Anchor it once when it is created, and anyone can check it afterwards.
Forged degrees, altered transcripts and invented certificates are easy to produce and slow to disprove. Checking one today means emailing a registrar and waiting days — so most employers simply do not check. Hikmaon turns that into a few clicks.
Because Hikmaon matches documents perceptually as well as by exact hash, a retouched scan does not quietly pass: a certificate with an edited grade, name or date comes back as a close match to the genuine record rather than a match, and the difference is shown as a percentage.
Prove your photos, videos and voice are yours, and have misuse found and taken down without hunting for it.
Defend against impersonation of your people and your brand, and verify media and documents before acting on them.
Timestamped, signed, independently checkable evidence of what a file was and when it existed.
Features
Detection that explains itself, ownership you can prove to a third party, and enforcement that never acts without your consent.
Perceptual hashing plus visual embeddings survive re-encoding, resizing and cropping, and report how close a copy is rather than a bare verdict.
Every registration is signed with Ed25519 and tied to its Hikmalayer transaction, so ownership can be verified independently of Hikmaon.
A politeness-respecting crawler indexes public media continuously and matches what it finds against your anchored fingerprints.
A multi-branch detector reading spatial texture, frequency artefacts and noise residuals, with calibrated probabilities instead of overconfident scores.
Videos are matched frame by frame with temporal alignment; audio by spectral fingerprints, so a clip lifted from a longer piece still matches.
Nothing is reported on your behalf unless you refuse consent. Every case carries the evidence report and is tracked to removal or rejection.
Intellectual property
The Hikmaon system and method is the subject of a filed patent application in the United Kingdom. The application covers blockchain-anchored authenticity verification, deepfake detection, ownership proof and automated rights-based notification.
Invented by Ayan Rao and developed at Bestower Labs. The application is pending; no grant is claimed.
About
Hikmaon exists because the people most harmed by synthetic media — students, professionals, public figures and ordinary users — are the least equipped to prove what is real. Hikmaon gives them infrastructure rather than advice: a verifiable record of what they created, and a system that watches for misuse on their behalf.
Hikmaon is currently working with selected users and institutions during its controlled beta phase. Pilot partners help evaluate the platform, provide structured feedback and contribute to the development of future capabilities.
Beta access is currently limited while the platform is validated. Selected pilot access may be provided during the beta programme. Commercial plans will be introduced as Hikmaon progresses beyond the beta phase.
Where Hikmaon is today
The technology described above is built and running — this is about the stage of the product and its route to market, not about whether it works.
A functioning platform in the hands of early users, researchers and selected institutions, running against real material.
Detection quality, monitoring scope and reporting workflows tightened against what beta and pilot feedback shows.
Wider commercial availability and institutional rollout once the platform has been validated in practice.
Join the beta as an early user, or bring your institution in as a pilot partner. Your feedback shapes the next version of Hikmaon.