01
Where SenseCrypt connects
Ping stays in the middle of the estate. Your service providers go on pointing at Ping, Ping goes on minting the sessions they read, and the connections, policies and reporting accumulated there are left alone. SenseCrypt joins at the top of that chain as an upstream identity provider.
What SenseCrypt assumes is the proof. On the mobile app methods the user faces the camera on the phone bound to them at enrollment, liveness and matching complete there, and Ping receives a signed token or assertion to map onto the record it already keeps. Upstream federation over OpenID Connect and SAML 2.0 has been standard Ping functionality for years, which is why this is a connection to configure rather than a component to run.
It is therefore a cheap experiment. Those connections are configured one at a time, so a single application or a single population can be moved and watched before anything else is touched.
Keep your existing identity hub
Your application inventory continues to depend on Ping and never on SenseCrypt. Nothing downstream has to learn a new endpoint, and no service provider configuration is rewritten.
Use SenseCrypt as an upstream provider
Ping redirects, the ceremony completes on the enrolled phone, and Ping folds the returned claims into the directory record it already holds for that person.
Use your existing federation protocol
OpenID Connect or SAML 2.0, chosen to match the rest of the Ping estate. Neither the ceremony on the device nor what the server retains depends on the wire format.
Pilot with one connection
Point the upstream provider at a single application or a single group. If that population does not take to it, detach the connection and they resume the previous method with nothing left to migrate.
02
Establish federation trust
Trust needs exactly two things from the SenseCrypt side: somewhere to send the user, and a way to check that the answer is authentic. Both are published documents, so nothing secret travels between administrators and no certificate is pasted into an email.
The rest of the effort is claim mapping, and it repays care. Ping builds its own user record and its own session out of the attributes you choose to send. An identifier that shifts the moment somebody updates an email address becomes account drift, usually discovered months later during an audit.
No biometric data is part of this exchange at all. On the two mobile app methods the check completes on the phone, and the federation carries only a signed statement that it completed for a named enrolled user. A webcam method that avoids the second device is offered to enterprise customers on a trusted customer network; contact sales@seventhsense.ai. This licensed on-premises method processes captures inside the customer deployment and does not add an enrolled-phone possession factor.
OIDC: one URL configures Ping
The discovery document hands Ping its endpoints and its verification keys in a single fetch, and it stays correct through a key rotation with no administrator involvement.
SAML 2.0: metadata both ways
Import the SenseCrypt IdP metadata on the Ping side, then send Ping's service provider metadata back the other way. Assertions are signed on issue, and Ping pins its trust to the published certificate.
Map the required claims
Standard claims or SAML attributes, together with the roles and permissions computed for that user. Ping lands them on its directory attributes, and your applications go on enforcing what they always enforced.
The boundary the biometric never leaves
On the two mobile app methods, neither a face image nor a biometric template leaves the handset. Ping is handed an authentication result, so your federation layer never holds biometric data of its own.
03
Integration steps
Everything here happens inside two administration consoles, which keeps the risk of the project in configuration rather than in infrastructure, and configuration can be reversed the same afternoon it was applied.
Run the first sign in on a real handset using a test identity, then inspect the record Ping wrote for that user. Mapping errors appear at exactly that point and nowhere sooner, and one account is a far cheaper place to find them than four hundred.
- Settle the protocol first: OpenID Connect or SAML 2.0 for the upstream link.
- In the SenseCrypt console, create a relying party record for Ping.
- In the Ping console, add SenseCrypt as an external identity provider.
- Give Ping the discovery URL for OIDC, or hand it the metadata file for SAML 2.0.
- Line the claims up with your Ping directory attributes.
- Attach the new provider to one connection or one population, not to everything.
- Enroll a test identity, sign in by face, then read back the record Ping created.
04
What changes in your deployment
Policies, connections and reporting survive intact, while the front of the flow loses its shared secret. That is the case for putting SenseCrypt above Ping rather than replacing anything: the blast radius of the decision is one connection.
It also puts attention on the part of the lifecycle attackers actually use. The 2023 MGM Resorts incident is the reference case: attackers reached the IT help desk by telephone, impersonated an employee, and the reported impact was approximately $100M. A project that hardens the front door while leaving account recovery to a persuasive phone call has relocated the attack rather than removed it.
SenseCrypt is deliberate about that path. Self-service face recovery does not exist, because a recovery route accepting a face without a possession proof would become the cheapest way in. Re-binding a device requires a possession proof: a one-time PIN to email, an SMS if a mobile number is recorded, or the CIBA email-possession check. Your service desk still needs a documented way to be certain who is on the line. The replacement must also prove a new device key; the account remains bound to the enrolled person through biometric-free, disposable face tokens, and the live face check is still required at sign-in.
No password for a phishing page
No password and no shared code appears at any point in the sign-in flow. The one-time PIN exists in a single place in the lifecycle, at device binding, which is a far smaller surface than a code issued on every login.
Where phishing resistance applies
The passkey path uses real FIDO2/WebAuthn passkeys (ES256) through a roaming authenticator app, and origin binding is the property that earns the phishing-resistant label there. The claim attaches to the path and not to the platform, because the QR path carries no origin binding. Phone sign-in also combines a live face check with a key bound to the enrolled device.
Testing and evaluation scope
Liveness detection tested by iBeta to ISO/IEC 30107-3 Levels 1 and 2 is presentation attack detection at the sensor. Face recognition is evaluated in the NIST Face Recognition Technology Evaluation under Seventh Sense's own name, participation since 2021. Evaluation and certification are different things, and blurring them is a habit worth refusing.
Withdrawal costs one connection
A federation is a single configured link. Where face login does not suit a population, delete the link: those users return to their previous method, with no data to migrate and no accounts to rebuild.
05
Frequently asked questions
Is SenseCrypt a Ping Identity partner?
No. What this page describes is standards-based federation between two products that both speak OpenID Connect and SAML 2.0. There is no partnership, no certification and no marketplace listing, which matters mainly because it means the configuration, the documentation and the testing all sit with your team.
Which protocol should we use for the upstream link?
Whichever your Ping estate already runs. OpenID Connect is the better default for a new connection, because discovery, PKCE (RFC 7636) and pushed authorization requests (RFC 9126) each do useful work on your behalf. SAML 2.0 is the pragmatic answer when the team already operates SAML connections and wants this one to look familiar during a night-time incident.
What does Ping receive after a face check?
A signed ID token or a signed SAML assertion naming the enrolled user, carrying the standard claims plus the roles and permissions computed for them. Ping validates the signature against the published key, maps those values onto its own user record, and then issues the session your applications consume.
Which device performs the match, and what is retained?
On the two mobile app methods every step stays on the phone the user enrolled: the capture, the liveness check and the comparison. IdP retains quantum-safe, sealed, biometric-free, disposable face tokens and their verifier challenges. Standard OIDC/SAML account records, device public keys, sessions, logs and encrypted tenant signing keys also persist. No face images or templates are retained in the documented phone flow. Licensed on-premises Webcam processes captures inside the customer deployment.
How is a lost or replaced phone handled?
The new device is bound again behind a possession proof: a one-time PIN to email, an SMS where the record holds a mobile number, or the CIBA email-possession check. Face recovery is never self-service, so the service desk procedure around that step belongs in the deployment plan and not in an afterthought. The replacement must also prove a new device key; the account remains bound to the enrolled person through biometric-free, disposable face tokens, and the live face check is still required at sign-in.
Does SenseCrypt hold a session alongside the Ping session?
Not for end users. Every authentication runs a fresh ceremony, and no SenseCrypt-side SSO session exists for end-user sign in, so the only session lifetime that matters is the one Ping issues. The SenseCrypt administration console does keep a browser session for the administrators who work in it.