Provenance · content authenticity on Aurigraph DLT · status as of 2026-10-20
Provenews is a blockchain-native content-provenance platform built on Aurigraph DLT. It answers a single, increasingly urgent question about any piece of media: where did this come from, and has it been altered since? Provenews binds a captured artifact — a photo, video, document, or article — to a cryptographic provenance chain anchored on-chain, so that any third party can verify the artifact's origin, integrity, and licensing state without trusting the platform that served it.
The platform expresses provenance as a 5-layer composite token that cryptographically binds content hash, capture metadata, creator credentials, device attestation, and a digital twin into a single verifiable object. The design verifies all five layers in parallel and returns a 0–100 authenticity score, with a latency target of under 100 milliseconds; the scoring service itself is design, not yet shipped code. Each verifiable claim is salted and each per-evidence step in a fact-trace chain is nonced, drawing that randomness from an attested entropy source — optionally sourced from quantum hardware through Aurigraph's shared QRNG agent, with a classical CSPRNG as the honest default. This paper describes the provenance model, the anchoring and verification protocol, the entropy design, and how Provenews integrates with the wider Aurigraph platform. Where a capability is production-grade it is stated as such; where it is directional it is labelled as direction.
Digital content is trivial to copy, trivial to alter, and — with generative models — trivial to fabricate from nothing. The institutions that historically vouched for a photograph or a document (a wire agency, a court clerk, a newsroom) no longer scale to the volume of media in circulation, and their trust is asserted rather than verifiable: a downstream consumer cannot independently check that a JPEG is the same one a photographer captured, or that a claimed source ever produced it.
This gap has concrete costs:
Traditional platforms address this with trust-based attestation, basic watermarks, and manual review measured in hours to days. Provenews replaces trust with verifiable cryptography: provenance is recorded once, at capture, and checked in milliseconds thereafter.
Provenews models authenticity as a composite token — a set of cryptographically bound layers, any one of which can be independently checked and none of which can be altered without invalidating the whole.
The lower layers (L1–L3) establish what the artifact is and where it came from; the upper layers (L4–L5) govern how it may be used. This paper focuses on L1–L3, the provenance core.
At capture, the raw artifact is hashed and the content is stored on IPFS, whose content-addressed identifiers are themselves derived from the bytes — so the storage address is an integrity check. The hash is the anchor for everything downstream: any modification to the artifact yields a different hash and therefore a different token identity.
The composite token — content hash, metadata digest, credential and attestation references — is anchored on Aurigraph DLT. Provenance records are expressed through Aurigraph's Ricardian / ActiveContract model, so each anchored artifact carries an immutable, timestamped contract hash on the ledger. The ledger stores commitments (hashes and references), not the media itself: the heavy bytes live on IPFS while the tamper-evident proof lives on-chain. This yields a complete provenance chain from capture through licensing, queryable via the platform's public-ledger surface — GET /api/v11/provenews/ledger/assets on dlt.aurigraph.io is live and public. Stated plainly: as of 2026-10-20 that endpoint still returns seeded demonstration assets rather than customer content; the removal of public demo data is in progress (AV11-3856..3859).
GET /api/v11/provenews/ledger/assets
dlt.aurigraph.io
Beyond static provenance, Provenews records a fact-trace: an ordered chain of provenance and verification events attached to an artifact or a claim. Each step — an ingestion, a re-verification, an evidence attachment, a licensing transition — is captured as a distinct nonced entry, so the trace forms a tamper-evident audit log rather than a single snapshot. Where provenance answers "is this the original?", the fact-trace answers "what has happened to it, in order, and can I replay that?" Each per-evidence nonce is drawn from the attested-entropy source described in §3 (platform purpose string provenews:facttrace-nonce), and each provenance-anchor claim is salted from the same source (provenews:factcheck-salt).
provenews:facttrace-nonce
provenews:factcheck-salt
Provenance salts and fact-trace nonces must be unpredictable: predictable salts allow precomputation attacks against anchored claims, and predictable nonces weaken the uniqueness guarantees of the audit chain. Provenews therefore sources this randomness from an attested entropy service rather than an ad-hoc local call.
Honest default. By default, entropy is drawn from a vetted classical CSPRNG, and the source of each batch is attested — the platform records which generator produced the bytes so a verifier can see the provenance of the randomness itself.
Optional quantum sourcing. Aurigraph operates a shared quantum-random-number-generator (QRNG) agent capable of drawing entropy from quantum hardware (an IBM quantum backend). When enabled and available, Provenews consumes this agent through named purpose strings — provenews:factcheck-salt for provenance-anchor salts and provenews:facttrace-nonce for per-evidence nonces — so entropy is namespaced per use and per consumer.
We do not claim live quantum entropy by default. The quantum path is a wired, opt-in capability, not the standing configuration: at the time of writing the shared QRNG agent is integrated but off by default, and the classical, attested fallback is what serves production. This is deliberate — the security of the provenance model rests on attested, unpredictable entropy, and the quantum backend is an upgrade to the entropy source, not a precondition for correctness. Any batch is labelled with its true origin (quantum or classical) so a downstream verifier is never misled about which was used.
The core promise of Provenews is that verification does not require trusting Provenews. A third party — a fact-checker, a court, an insurer, a downstream publisher — can independently confirm a provenance claim:
The design performs this as a parallel 5-layer check and returns a 0–100 authenticity score — with classifications spanning Authentic, Probably Authentic, Suspicious, Probably Fraud, and Fraud — with a design-target latency of under 100 ms per verification (not measured on production). The scoring service is not yet implemented in the V12 tree, and no latency has been measured. Crucially, steps 1–4 are reproducible by anyone holding the artifact and read access to the public ledger, independent of the scoring service.
V12
Provenews is a first-class product on Aurigraph DLT rather than a standalone chain, and inherits the platform's provenance, identity, and settlement primitives:
This integration means Provenews does not reinvent consensus, identity, or settlement — it composes them. The provenance chain, the credential system, and the ledger anchoring are shared Aurigraph capabilities that Provenews specializes for media authenticity.
Provenews reframes content authenticity from an act of trust into an act of verification. By binding content hash, capture metadata, creator credentials, and device attestation into a cryptographically coherent composite token, anchoring it on Aurigraph DLT, and layering a nonced fact-trace on top, it produces provenance claims that any third party can check independently and in milliseconds. Its entropy is attested — honestly classical by default, optionally quantum-sourced through Aurigraph's shared QRNG agent — so the unpredictability underpinning the model is itself accountable. As generative media makes "seeing is believing" obsolete, the durable primitive is verifiable provenance; Provenews aims to make that primitive routine, fast, and platform-independent.
ProvenewsLedgerResource
/api/v11/provenews/ledger/assets
Explore: About · Products · Features · Whitepapers · Sustainability · Tokenized carbon credits · Real-world asset tokenization · SDK · Blog · Plain-text site index · Sitemap · Contact: contact@aurigraph.io