A trust list you pay to join, and a cert you rent per signer, is provenance priced backwards.
Making a signer publicly recognized today means a ~$289/yr commercial C2PA certificate, renewed per signer, behind a list a small coalition curates — and an off-list signer displays "unknown source" no matter how honest it is. There is no Let's Encrypt for provenance. That's a toll and a gate on who may be trusted — exactly the wrong shape when synthetic content is already the majority and most of it is unsigned.
whisper verify --trustless costs nothing and needs no account — our own API is not in the trust path.
A per-cert toll that climbs with every signer you onboard — and resets when a CA fails — isn't a price. It's a tax on being trusted.
Two curves. One rises with every stringer, desk, camera and agent you need to make publicly verifiable — a fresh ~$289/yr certificate each, and a whole re-cert when a CA is compromised. The other is a flat line you set once, forecast for years, and that turns any signer public in your own DNS with no list to join.
Per-signer identity, not per asset
Priced to the thing you actually govern — the signer, be it a desk, a stringer, a camera model or an agent — so signing a million assets, or a viral verification spike, never moves the invoice. Onboard a new signer and it's one flat line, not another ~$289/yr cert.
Verify & who-verified are never metered
Keyless whisper verify --trustless is free forever — our API isn't in the trust path — and the graph and op:lookups run unmetered on the keyed tiers. No per-verification tax means you never ration a hunt while an impostor cert keeps circulating.
No CA fee, no list to join
A flat per-signer-identity line replaces the commercial C2PA cert and the pay-to-play list. Publish the signer's EE cert under your own DNSSEC/DANE record and it's self-verifying — no gatekeeper, no phone-home, no "unknown source" penalty on an honest off-list signer.
Start keyless and free. Prove it on a handful of signers. Roll it across the publisher — flat the whole way.
POC → pilot → enterprise, exactly the path a provenance program buys on. Every tier speaks the same address-is-the-signer primitive; you're only widening how many signers it covers, never re-platforming — and the manifest stays C2PA's the whole time.
Free to verify
$0
Keyless forever. A handful of signer identities to prove the flow.
The keyless half of the platform — trustless and anchored at the IANA root, our API never in the path — plus a proof-of-concept set of signers:
- ✓
whisper verify --trustlessany signer — no key, no CA - ✓ Resolve and reverse-resolve a signer
/128, read its RDAP - ✓ Provision a handful of signer identities — DANE-anchored, from the C2PA signer cert serial they carry
A desk or a product line
Flat engagement
A newsroom desk or one product line's signing fleet, time-boxed, one flat price.
Everything in POC, keyed to a defined signer count so a program owner can prove value before the board:
- ✓ Full attribution graph — unmetered during the pilot
- ✓ "Who verified my content" analytics (
op:lookups) — the empty quadrant, as a stream - ✓ Machine-readable feed into your SIEM: Splunk & Microsoft Sentinel today (STIX 2.1 / TAXII on the roadmap)
- ✓ Merkle transparency-log evidence · EU AI Act Art.50 audit export
Flat per-signer / year
Signer quote
One rate, quoted to your signer count. It doesn't move.
The whole publisher or the whole device fleet — every signer, on your own terms:
- ✓ Identity, attribution graph, who-verified analytics and egress governance, fleet-wide
- ✓ Per-unit
op:revoke— kill one compromised signer worldwide at DNS-TTL, never a fleet-wide reset - ✓ On-prem or your own tenant — GDPR, data residency and source-protecting pseudonymity by construction
- ✓ Enterprise support and SLA
# keyless — verify any signer, free forever, our API not in the trust path
$ whisper verify --trustless newsroom.example.press
✓ DNSSEC chain valid to the IANA root
✓ DANE-EE (TLSA) matches the C2PA signer EE cert · serial 03:a8:f1:9c:…
✓ RDAP: registered under AS219419 · 2a04:2a01::/32
signer: VERIFIED — no central Trust List, no CA fee
# provision a signer from the C2PA signer cert serial it already carries (device_id)
# — a first-class --signer arg is on the roadmap; the serial is the socket today
$ curl -s https://graph.whisper.security/api/query -H "X-API-Key: whisper_live_xxx" --data-urlencode "q=CALL whisper.agents({op:'connect', args:{tier:'wireguard',
identity_public_key:'<base64 SPKI of the signer key>',
device_id:'03A8F19C4E7B2D01'}})" # device_id = the signer cert serial
→ signer 2a04:2a01:5e0::c2 DNSSEC + DANE-EE live · in the transparency log
# "who verified my content" — the write-only quadrant, now a stream (unmetered)
$ whisper lookups 2a04:2a01:5e0::c2
1,284 verifications · 37 countries · 6 platforms — last 24h
⚠ 3 lookups resolved a signer serial you never issued → possible impostor certWhy a quote, not a sticker. A signer price is one number, but the right number depends on how many signers you publish, on-prem vs your own tenant, and the standards evidence you need — so we quote it flat and in writing, and it holds for the term. No per-cert renewals, no usage true-ups, no surprise line at renewal. Get a signer quote →
The same platform, at three widths. Nothing behind the paywall is the ability to verify.
The keyless verification a reader, a fact-checker, a platform or a regulator needs to resolve a signer's identity is free at every tier — on principle. The keyed tiers widen coverage and feed your stack; they never gate the public check. And we never create the C2PA manifest or embed a watermark — we anchor the signer those manifests already reference.
| Capability | POC | Pilot | Enterprise |
|---|---|---|---|
Trustless verify / resolve / RDAP (whisper verify --trustless) | ✓ | ✓ | ✓ |
| Trace a signer /128 to the identity behind it (reverse-DNS + RDAP) | ✓ | ✓ | ✓ |
| Signer /128 identities + DANE-anchored signer certs (from the C2PA cert serial) | a handful | desk / line | fleet-wide |
Full attribution graph (identify, origins, walk, history, Cypher) | — | unmetered | unlimited |
"Who verified my content" analytics (op:lookups) | — | ✓ | ✓ |
Egress governance for signing agents (policy, firewall, budget, op:logs) | — | desk / line | fleet-wide |
Per-unit revocation at DNS-TTL (op:revoke) | your signers | desk / line | fleet-wide |
| Merkle transparency log — issuance & revocation audit trail | public read | ✓ | ✓ |
| SIEM feed: Splunk & Microsoft Sentinel connectors today (STIX 2.1 / TAXII roadmap) | — | ✓ | ✓ |
| EU AI Act Art.50 / ISO 22144 (draft) evidence export (JSON export on roadmap) | — | ✓ | ✓ |
| On-prem / own tenant (data residency, GDPR, pseudonymity) | — | — | ✓ |
| Enterprise support & SLA | — | pilot support | ✓ |
| Per-signer commercial-CA fee (~$289/yr each) | never | never | never |
| Metered by usage (per asset signed / per verification / per seat) | never | never | never |
One honest note on standards. DNSSEC/DANE anchoring is a legitimate complementary signer-identity ecosystem — C2PA trust anchors are pluggable config inputs to a validator, and a DANE anchor is surfaced through a CAWG identity assertion — but it is not (yet) a formally recognized C2PA conformance trust anchor. We're pursuing that formal-recognition track through CAWG and the standard; we don't price it as done. What ships today makes any signer publicly verifiable and revocable in open DNS — additive to C2PA, never instead of it.
The ROI isn't a promise — it's costs the flat line takes off your books, and one it puts back on.
A predictable figure is only half the case. The other half is what it removes — a per-signer CA toll, a fleet-wide cert reset, audit effort, re-platform risk — and the one new line it can add: "who verified my content," a stream no incumbent can sell.
No ~$289/yr CA fee, no gatekeeping
Every stringer, desk, camera model and agent becomes a publicly-verifiable signer via your own DNSSEC/DANE — no per-cert commercial-CA fee, no Trust-List slot to buy, no "unknown source" penalty on an honest off-list signer. The "no Let's Encrypt for provenance" gap, closed at a flat line.
One revoke, not a fleet-wide cert reset
A compromised signer key is revoked worldwide at DNS-TTL in one call — per unit. When a camera vendor's authenticity service was suspended and its entire set of device certs had to be revoked in 2025, that was a fleet-wide reset; per-unit revocation bounds the blast radius to one leaf key, never a shared root — the shared-CA failure mode is structurally removed.
"Who verified my content" — a new line, not a cost
C2PA verification needs no network call, so a signer normally has zero visibility into who checked their content. Anchor the signer in DNS and op:lookups turns that empty write-only quadrant into a stream — an impersonation early-warning and a sellable B2B analytics product no incumbent can offer.
Attribution & verify are never metered
Run identify, walk, history and Cypher — and keyless verify — as hard as an incident demands. No per-verification tax means a fact-checker triaging a viral deepfake, or a platform verifying at ingestion, never rations the check while an impostor cert keeps working.
Audit effort you don't repeat
Every mint and every revoke lands in a public, append-only RFC 6962 Merkle transparency log, Ed25519-signed and Bitcoin-anchored via OpenTimestamps — a non-repudiable issuance trail that evidences the "cryptographic methods for proving provenance" EU AI Act Recital 133 enumerates. A byproduct of the tool, not a consulting line. Honest status: tamper-evident today, independent witnessing is the next step.
A vendor that will still be here
Real routable address space (AS219419), run by people who ran the internet's regional address registry and operated one of its root DNS servers. Your signer identities are DNSSEC/DANE objects anyone can verify independently — longevity is the cheapest line in any TCO.
A pricing model can be a trap. Ours isn't — and we'll tell you exactly what it is and isn't.
If verification is metered, an adversary can run up your bill and a defender rations their own check. If trust is gate-kept, honest signers pay a toll to be believed. We priced both failure modes out — and we're candid about the standards line we haven't crossed yet.
"If verification is metered, does my bill spike when my content goes viral — or when someone floods checks to run it up?"
Neither. Keyless verify is free forever and our own API is not in the trust path, so there's no per-verification line for a viral moment to inflate or an adversary to weaponize. Attribution and op:lookups are unmetered on the keyed tiers. The only variable on your invoice is how many signers you publish.
"Is this just a second CA fee stacked on top of the C2PA cert I already buy?"
No — it replaces the toll, not adds one. There's no per-cert ~$289/yr commercial-CA fee and no Trust-List slot to buy. You derive the signer identity from the key the signer already holds and publish its EE cert under your own DNSSEC/DANE — self-verifying, no gatekeeper. It makes an off-list signer publicly verifiable; it doesn't make you buy a second certificate to do it.
"Are you charging me for something you're claiming is 'C2PA-approved'? Because I'll check."
We're not — and here's the honest line. DNSSEC/DANE anchoring is a legitimate complementary identity ecosystem (C2PA trust anchors are pluggable validator inputs, surfaced through a CAWG identity assertion), but it is not yet a formally recognized C2PA conformance trust anchor. We're pursuing that recognition through CAWG and the standard, and we price only what ships: a publicly-verifiable, revocable signer identity in open DNS — additive to your C2PA manifest, never a replacement for it.
Straight answers, before the call.
What exactly is metered?
Nothing by usage. You pay a flat rate per signer identity, per year. No per-asset-signed charge, no per-verification fee, no per-analyst seat, no per-cert commercial-CA fee, no data-egress bill. The only variable is how many signers the program covers.
Can I try it without procurement?
Yes — keyless verify is free and needs no account. Run whisper verify --trustless against any signer today; a POC lets you provision a handful of signer identities from the C2PA cert serial they carry, no commitment.
What happens as I onboard more signers?
The per-signer rate holds; the total scales linearly and predictably with signer count, quoted in writing for the term. No per-cert ~$289/yr tax, no renewal surprise, and an off-list signer never costs extra to make public.
Is this on top of my C2PA cost?
It's additive to C2PA / Content Credentials — it anchors the same signer the manifest already references. We don't create the manifest or embed a watermark; we make the signer publicly verifiable and revocable. The Splunk connector feeds your existing SIEM, not a second console.
On-prem or hosted?
Either. The Enterprise tier runs on-prem or in your own tenant, so the graph and per-signer logs stay where your regulator needs them. Org- or domain-level anchoring with pseudonymity protects journalists and whistleblowers — GDPR and data residency by construction, at no metered premium.
What if I stop?
Your signer identities are DNSSEC/DANE objects you can verify independently, transparency-log proofs stay valid, and evidence exports are open formats (CEF, ECS; STIX and per-sector JSON on the roadmap). There's no proprietary lock on your own attestations or your compliance record.
Flat depth on top of Content Credentials — it doesn't replace the manifest, it makes the signer public.
You already emit C2PA manifests and, maybe, pay a commercial CA to be a recognized signer — keep the manifest; it carries the edit history and the AI-generated assertion your Art.50 mark needs. Where a per-cert CA taxes each signer and gate-keeps who may be trusted, and a hypothetical per-verification cloud makes the bill unforecastable, a flat per-signer-identity line adds the two things no one else owns — a publicly-verifiable, revocable signer in open DNS and "who verified my content" analytics — without a meter, a gatekeeper, or a second silo.
| Trust / pricing model | Forecastable & flat? | Publicly verifiable off-list, no CA fee? |
|---|---|---|
| Per-cert commercial C2PA CA (~$289/yr, gate-kept list) | partly — scales per signer | no — needs the list |
| Self-signed / off-list cert | $0 | no — shows "unknown source" |
| Whisper — flat per-signer-identity / year (DNSSEC/DANE) | yes | yes |
It makes the Content Credentials, watermarking and verification tooling you already run sharper, as a machine-readable feed — not a thing they compete with. See the full comparison →
One flat number. Every signer, publicly verifiable.
Keyless verify is free forever — start there, no account. When you're ready, a signer quote is one flat per-signer-identity/year figure you can forecast and defend: no CA fee, no Trust-List slot, no meter, no surprise at renewal.
Or run whisper verify --trustless right now — it costs nothing.