1. Resources
  2. Detection rules
  3. MFA Device Registered or Removed for a PingOne User

MFA Device Registered or Removed for a PingOne User detection rules

An attacker who controls a PingOne account, or an administrator role over it, pairs an MFA device of their own with the user or removes the user's device, so that later sign-ons pass the second factor.

Platform
Ping Identity
Severity
Medium
Log source
PingOne audit activities
Last updated
Rule formats
Sigmaand queries for: SplunkMicrosoft Sentinel
What an attacker needs
A session of the target user where devices can be managed, or a role that can manage the MFA devices of other users.
Impact
The attacker's device satisfies MFA for the account for as long as it stays registered, also after the password is reset.
Response
Delete the device the user does not recognize, end the user's sessions, reset the password and re-enroll MFA through a verified channel.
Mitigation
Require a fresh MFA step before a new device is paired, and notify users of every new device with a link to report it.
Action types
FRAUD.REPORTEDDEVICE.FRAUD_REPORTEDDEVICE.CREATEDDEVICE.ACTIVATEDDEVICE.DELETEDDEVICE.UNLOCKEDDEVICE.UNBLOCKED

How the attack works

In PingOne an MFA method is a device that belongs to a user: an email address, a phone number for SMS or voice, an authenticator app, a FIDO2 authenticator, a mobile application or a hardware token. A user can hold several, by default up to 5. Whoever adds a device to an account owns a second factor for it. An attacker with a session of the user pairs their own authenticator app or phone number. An attacker with an administrator role does not need the user at all: Ping documents that an actor who creates a device on behalf of another user, typically a worker application, can create it with the status ACTIVE, which means the device is pre-paired and the user is not required to activate it before it is used for authentication.

The opposite change serves the same goal. Deleting a user's strong device leaves the weaker ones, since the next active device in the order becomes the default, and releasing a device that was locked or blocked gives a guesser or a stolen phone a new series of tries. PingOne records these operations as DEVICE.CREATED, DEVICE.ACTIVATED, DEVICE.DELETED, DEVICE.UNLOCKED and DEVICE.UNBLOCKED audit activities. The user can also speak up: the MFA policy can notify a user that a new authentication method has been added to their account (newDeviceNotification), and Ping's New Device Paired notification template can carry a link for reporting fraudulent pairing attempts. The event list holds two activities described as Fraud Reported, FRAUD.REPORTED and DEVICE.FRAUD_REPORTED. The page follows Ping's documentation of these operations: no public incident report describes them for PingOne at the audit event level.

Log source: the rules read PingOne audit activities. Include the Device and Fraud event types in the PingOne webhook that feeds the SIEM, or poll the Audit Activities API, for every environment. Device activities are written for every enrollment, so the rules on created, deleted and released devices are for review and for context, and the fraud report is the one to alert on.

What an attacker needs

  • A session of the target user in an application that allows device management, or a role that can manage MFA devices of other users. Ping names the Identity Data Admin role for operations on device resources.
  • For a pairing by the user's own session, an MFA policy that allows the device type the attacker wants to add.

Impact

The attacker's device satisfies MFA for the account for as long as it stays registered, also after the password is reset. A removed or weakened set of devices lets a known password and an interceptable passcode suffice. For an administrator account this is durable access to the PingOne environment itself.

Detection rules

PingOne User Reported Fraud

Detects the audit activities FRAUD.REPORTED and DEVICE.FRAUD_REPORTED, both of which Ping's event list describes as Fraud Reported. Ping documents where such a report comes from: the MFA policy can notify a user when a new authentication method has been added to their account (newDeviceNotification), and the New Device Paired notification template supports a variable that inserts a link for reporting fraudulent pairing attempts. A report therefore is the user's own statement that a device was paired with their account by someone else, which means that an attacker held a session or the password of the user, or a role that can pair devices for others. The event list gives no further description of the two activities, so confirm with one test report which of them your environment writes and what it names in its resources. Few signals in this log come from a person, which is why the level is high.

LevelHighStatusExperimentalLog sourcePingOne audit activities
title: PingOne User Reported Fraud
id: ba90b7b8-e922-4c12-879e-8aa955d7c789
status: experimental
description: >-
  Detects the audit activities FRAUD.REPORTED and DEVICE.FRAUD_REPORTED, both
  of which Ping's event list describes as Fraud Reported. Ping documents where
  such a report comes from: the MFA policy can notify a user when a new
  authentication method has been added to their account
  (newDeviceNotification), and the New Device Paired notification template
  supports a variable that inserts a link for reporting fraudulent pairing
  attempts. A report therefore is the user's own statement that a device was
  paired with their account by someone else, which means that an attacker
  held a session or the password of the user, or a role that can pair devices
  for others. The event list gives no further description of the two
  activities, so confirm with one test report which of them your environment
  writes and what it names in its resources. Few signals in this log come
  from a person, which is why the level is high.
references:
  - https://attack.mitre.org/techniques/T1098/005/
  - https://developer.pingidentity.com/pingone-api/platform/reference/audit-reporting-events.html
  - https://developer.pingidentity.com/pingone-api/platform/notifications/notifications-templates/index.html
  - https://developer.pingidentity.com/pingone-api/mfa/device-authentication-policy/index.html
author: Forestall
date: 2026-11-18
tags:
  - attack.persistence
  - attack.privilege-escalation
  - attack.t1098.005
logsource:
  product: pingone
  service: audit
detection:
  selection:
    action.type:
      - FRAUD.REPORTED
      - DEVICE.FRAUD_REPORTED
  condition: selection
falsepositives:
  - A user who follows the report link by mistake, or who forgot that they paired a new phone or that the help desk paired one for them. Ask the user before you close the report, and check who created the device.
level: high

False positives

  • A user who follows the report link by mistake, or who forgot that they paired a new phone or that the help desk paired one for them. Ask the user before you close the report, and check who created the device.

PingOne MFA Device Created or Activated

Detects a new MFA device for a PingOne user (audit activities DEVICE.CREATED and DEVICE.ACTIVATED, described by Ping as Device Created and Device Activated). A device is an MFA method of the user, and whoever pairs one holds a second factor for the account. Ping documents two paths. A user who creates a device for themselves gets the status ACTIVATION_REQUIRED and has to activate it. An actor who creates the device on behalf of another user, typically a worker application, can create it as ACTIVE, which is the default in that case: the device is pre-paired and the user is not required to activate it. The rule matches both activities because Ping does not document whether a pre-paired device also writes DEVICE.ACTIVATED. Every enrollment fires it, so use it for review and compare the actor (actors.user.name or actors.client.name) with the user named in the activity's resources.

LevelLowStatusExperimentalLog sourcePingOne audit activities
title: PingOne MFA Device Created or Activated
id: cfcf41dc-e5ac-4b1c-855f-5586785b30e9
status: experimental
description: >-
  Detects a new MFA device for a PingOne user (audit activities DEVICE.CREATED
  and DEVICE.ACTIVATED, described by Ping as Device Created and Device
  Activated). A device is an MFA method of the user, and whoever pairs one
  holds a second factor for the account. Ping documents two paths. A user who
  creates a device for themselves gets the status ACTIVATION_REQUIRED and has
  to activate it. An actor who creates the device on behalf of another user,
  typically a worker application, can create it as ACTIVE, which is the
  default in that case: the device is pre-paired and the user is not required
  to activate it. The rule matches both activities because Ping does not
  document whether a pre-paired device also writes DEVICE.ACTIVATED. Every
  enrollment fires it, so use it for review and compare the actor
  (actors.user.name or actors.client.name) with the user named in the
  activity's resources.
references:
  - https://attack.mitre.org/techniques/T1098/005/
  - https://developer.pingidentity.com/pingone-api/platform/reference/audit-reporting-events.html
  - https://developer.pingidentity.com/pingone-api/mfa/users/mfa-devices/index.html
author: Forestall
date: 2026-11-18
tags:
  - attack.persistence
  - attack.privilege-escalation
  - attack.t1098.005
logsource:
  product: pingone
  service: audit
detection:
  selection:
    action.type:
      - DEVICE.CREATED
      - DEVICE.ACTIVATED
  condition: selection
falsepositives:
  - Normal enrollment. New users, users with a new phone and rollouts of a new method produce these activities all day. Add a filter on actors.client.name for the enrollment applications you have reviewed, or alert only for administrator accounts.
  - Help desk staff who pair a device for a user on request. Confirm against the ticket.
level: low

False positives

  • Normal enrollment. New users, users with a new phone and rollouts of a new method produce these activities all day. Add a filter on actors.client.name for the enrollment applications you have reviewed, or alert only for administrator accounts.
  • Help desk staff who pair a device for a user on request. Confirm against the ticket.

PingOne MFA Device Deleted

Detects an MFA device being deleted from a PingOne user (audit activity DEVICE.DELETED, described by Ping as Device Deleted). Removing a user's strong device changes what the next sign-on asks for: Ping documents that when the default device is deleted, the second active device in the order becomes the default, so a deleted FIDO2 authenticator can leave an email or SMS passcode as the factor in use. An attacker also deletes the user's own device after pairing theirs, to keep the user from noticing prompts. The activity does not say which type of device was removed or what the user has left: read the user's devices during triage. The rule fires on every deletion, so use it for review.

LevelLowStatusExperimentalLog sourcePingOne audit activities
title: PingOne MFA Device Deleted
id: 6a9f800c-f99c-410c-adad-3d5b2442af72
status: experimental
description: >-
  Detects an MFA device being deleted from a PingOne user (audit activity
  DEVICE.DELETED, described by Ping as Device Deleted). Removing a user's
  strong device changes what the next sign-on asks for: Ping documents that
  when the default device is deleted, the second active device in the order
  becomes the default, so a deleted FIDO2 authenticator can leave an email or
  SMS passcode as the factor in use. An attacker also deletes the user's own
  device after pairing theirs, to keep the user from noticing prompts. The
  activity does not say which type of device was removed or what the user has
  left: read the user's devices during triage. The rule fires on every
  deletion, so use it for review.
references:
  - https://attack.mitre.org/techniques/T1556/006/
  - https://developer.pingidentity.com/pingone-api/platform/reference/audit-reporting-events.html
  - https://developer.pingidentity.com/pingone-api/mfa/users/mfa-devices/index.html
author: Forestall
date: 2026-11-18
tags:
  - attack.defense-impairment
  - attack.persistence
  - attack.credential-access
  - attack.t1556.006
logsource:
  product: pingone
  service: audit
detection:
  selection:
    action.type: DEVICE.DELETED
  condition: selection
falsepositives:
  - Users who replace a phone or remove a method they no longer use, and help desk staff who remove a lost device on request.
  - Clean-up by PingOne. Ping documents that devices left in the ACTIVATION_REQUIRED status for 24 hours are deleted from the system. Whether that writes this activity is not documented, so check the actor before you exclude anything.
level: low

False positives

  • Users who replace a phone or remove a method they no longer use, and help desk staff who remove a lost device on request.
  • Clean-up by PingOne. Ping documents that devices left in the ACTIVATION_REQUIRED status for 24 hours are deleted from the system. Whether that writes this activity is not documented, so check the actor before you exclude anything.

PingOne MFA Device Unlocked or Unblocked

Detects an MFA device being released (audit activities DEVICE.UNLOCKED and DEVICE.UNBLOCKED, described by Ping as Device Unlocked and Device Unblocked). Ping documents two restrictions on a device. A lock is set automatically after too many unsuccessful passcode tries or too many declined or ignored push notifications, and a block is set through the block operation of the API. Both can be lifted through the API by an actor with the permission to update devices. A release gives whoever caused the lock a new series of tries, and unblocking a device that was blocked because it was lost or stolen puts it back into use. The request often reaches the help desk by phone, which makes it a target for social engineering. Ping does not document whether a lock that expires by itself writes DEVICE.UNLOCKED, so read the actor first.

LevelLowStatusExperimentalLog sourcePingOne audit activities
title: PingOne MFA Device Unlocked or Unblocked
id: c14ac45c-6b55-4a73-b8f1-4d631ef7bf06
status: experimental
description: >-
  Detects an MFA device being released (audit activities DEVICE.UNLOCKED and
  DEVICE.UNBLOCKED, described by Ping as Device Unlocked and Device
  Unblocked). Ping documents two restrictions on a device. A lock is set
  automatically after too many unsuccessful passcode tries or too many
  declined or ignored push notifications, and a block is set through the
  block operation of the API. Both can be lifted through the API by an actor
  with the permission to update devices. A release gives
  whoever caused the lock a new series of tries, and unblocking a device that
  was blocked because it was lost or stolen puts it back into use. The
  request often reaches the help desk by phone, which makes it a target for
  social engineering. Ping does not document whether a lock that expires by
  itself writes DEVICE.UNLOCKED, so read the actor first.
references:
  - https://attack.mitre.org/techniques/T1556/006/
  - https://developer.pingidentity.com/pingone-api/platform/reference/audit-reporting-events.html
  - https://developer.pingidentity.com/pingone-api/mfa/users/mfa-devices/index.html
author: Forestall
date: 2026-11-18
tags:
  - attack.defense-impairment
  - attack.persistence
  - attack.credential-access
  - attack.t1556.006
logsource:
  product: pingone
  service: audit
detection:
  selection:
    action.type:
      - DEVICE.UNLOCKED
      - DEVICE.UNBLOCKED
  condition: selection
falsepositives:
  - Help desk staff who release a device for a user who locked it with mistyped passcodes. Confirm against the ticket and with the user through another channel.
  - Locks that end after the lock duration of the MFA policy, if your environment writes the activity for them.
level: low

False positives

  • Help desk staff who release a device for a user who locked it with mistyped passcodes. Confirm against the ticket and with the user through another channel.
  • Locks that end after the lock duration of the MFA policy, if your environment writes the activity for them.

Detection logic

Patterns that need more than one event, a baseline or knowledge of your own environment. Each is a hypothesis to build and tune in your SIEM, not a ready rule.

MFA device registered shortly after a password recovery or reset of the same user

Detection hypothesisSequence of events

An attacker who takes over an account through password recovery or a reset by the help desk needs a second factor next. A new device for a user within a short time after that user's password was recovered, reset or set is the usual order of an account takeover, and much rarer than either activity alone.

  1. Find PASSWORD.RECOVERY, PASSWORD.RESET or PASSWORD.SET activities and note the user they concern, named in resources.name where resources.type is USER.
  2. Look for DEVICE.CREATED or DEVICE.ACTIVATED for the same user within the next 60 minutes.
  3. Compare source.ipAddress of the password activity and of the device activity with the addresses the user normally signs on from.
  4. Check whether an older device of the user was deleted (DEVICE.DELETED) in the same period.

EventsPASSWORD.RECOVERYPASSWORD.RESETPASSWORD.SETDEVICE.CREATEDDEVICE.ACTIVATEDDEVICE.DELETED

LimitsA new employee and a user with a new phone produce the same order, because onboarding sets a password and then enrolls a device. Ping's event list does not say which password activity is written for a change by the user and which for one by an administrator. The window is a starting point.

MFA device added for a user by an actor that is not on the list of enrollment identities

Detection hypothesisNeeds your own lists

Devices are normally added by the user, or by a small set of identities such as the help desk and an enrollment worker application. A device that some other administrator or application creates for a user, above all with the pre-paired status, is the administrative form of this attack and stands out against that short list.

  1. Collect DEVICE.CREATED activities and compare the actor (actors.user.id, or actors.client.id when no user is the actor) with the user the device belongs to (resources.id where resources.type is USER).
  2. Drop the activities where the actor is the user, and those where the actor is on your reviewed list of help desk administrators and enrollment applications.
  3. For what remains, read the device through the API and check whether its status was ACTIVE from the start, and whether the user signed on with it afterwards.

EventsDEVICE.CREATED

LimitsIt needs your own list of the identities that enroll devices, and the list has to be kept current. Ping's event list does not state which entry of the resources array is the user and which the device, so check the comparison against a few known enrollments first.

MFA device registered shortly after a high-risk evaluation of the same user

Detection hypothesisSequence of eventsNeeds a baseline

PingOne Protect rates sign-ons. An attacker who signs on with a stolen password from an unusual address or device is likely to be rated HIGH or MEDIUM, and the next step of a takeover is a second factor of their own. A new device for a user within a short time of a high-risk evaluation of that user joins two signals that each mean little alone.

  1. Find RISK_EVALUATION.CREATED activities whose _embedded.riskResult.level is HIGH or MEDIUM. Keep the user (_embedded.riskEvent.userId, _embedded.riskEvent.username) and the address (_embedded.riskEvent.ip), attributes that Ping's event list names for this activity.
  2. Look for DEVICE.CREATED or DEVICE.ACTIVATED for the same user within the next 60 minutes, the user being the entry of resources where resources.type is USER.
  3. Compare source.ipAddress of the device activity with _embedded.riskEvent.ip and with the addresses the user signed on from before. The same address as the risky evaluation, or one the user has not used, is the case to act on.

EventsRISK_EVALUATION.CREATEDDEVICE.CREATEDDEVICE.ACTIVATED

LimitsNeeds PingOne Protect with risk evaluations requested for sign-ons. A user who travels and pairs a new phone produces the same pair. Ping's event list gives these attributes no description, so check them against one evaluation read through the Risk Evaluations API, which documents the levels HIGH, MEDIUM and LOW. The 60 minutes are a starting point.

MFA device registered shortly after the user's Active Directory password was reset

Detection hypothesisJoins two logsSequence of events

Where users' passwords live in Active Directory and PingOne checks them through a gateway, a help desk reset is made in the directory and recorded by the domain controllers, so the PingOne password activities in the first entry above do not show it. The takeover order is the same: someone else resets the password, then a new device appears. Joining the reset event of the domain controllers with the device activity in PingOne shows it.

  1. On the domain controllers, collect Security event 4724 (An attempt was made to reset an account's password), where SubjectUserName is the account that made the reset and TargetUserName the account that was reset. Keep the resets where the two differ.
  2. In PingOne, look for DEVICE.CREATED or DEVICE.ACTIVATED within 60 minutes for the user whose name matches TargetUserName, the user being the entry of resources where resources.type is USER.
  3. Normalise the names before the join: the Windows event logs the account name without the domain, while PingOne user names are often a user principal name or an email address.
  4. Compare source.ipAddress of the device activity with the addresses the user normally signs on from, and confirm the reset with the user through a separate channel.

Events4724DEVICE.CREATEDDEVICE.ACTIVATED

LimitsNeeds the Security logs of all domain controllers in the same SIEM, with User Account Management auditing on. Onboarding, and a user who calls the help desk for a forgotten password and then sets up a new phone, produce the same pair. Changes users make themselves (event 4723) are left out. The 60 minutes are a starting point.

Triage

  • Identify the actor (actors.user.name or actors.client.name) and the user the device belongs to, named in the activity's resources. A device added by anyone but the user needs an approved request.
  • Read the device through the API or the console. Check its type, its status, and for SMS, voice and email devices the phone number or address, and compare them with what you know about the user.
  • Look at what happened to the account shortly before, such as a password recovery or reset, failed password or passcode checks, or a sign-on from a new address (source.ipAddress).
  • For a device of the type OATH token, check who added the token to the environment (OATH_TOKEN.CREATED) and when.
  • For a deleted device, list the devices the user has left and whether a new one was added by the same actor in the same session.
  • For a fraud report, ask the user what they received and what they did not do, and treat the device named in the report as hostile until shown otherwise.

Response

Once the attack is confirmed:

  1. Delete the device the user does not recognize, and block it first if you need to keep its record for the investigation.
  2. End the user's sessions and reset the password, because the pairing shows that the first factor or a session was in the attacker's hands.
  3. Re-enroll the user's MFA with a method verified in person or through a channel that does not depend on the account.
  4. If an administrator or worker application made the change, suspend it, remove the role that allows it and review every device it created, deleted or released.
  5. Review what the account did since the device was added, above all changes to roles, applications, secrets and other users' devices.

Mitigation

  • Require a fresh MFA step with an existing device before a new device can be paired, in the flows and applications where users manage their devices.
  • Keep newDeviceNotification of every MFA policy on a value that notifies the user, and keep the report link in the New Device Paired template, so that users can report a pairing they did not make.
  • Limit the Identity Data Admin role and custom roles with device permissions, and review the worker applications that hold them.
  • Lower the maximum number of devices per user in the MFA settings to what users need, and allow only the device types you want in each MFA policy.
  • Review the devices of administrators regularly and remove those the owner does not recognize.

Frequently asked questions

Can a device be added to a PingOne user without the user taking part?

Yes, by an actor with the right role. Ping documents that when a device is created on behalf of another user, typically by a worker application, its status can be `ACTIVE`, which is also the default in that case. The device is then pre-paired and the user is not required to activate it. A user who creates a device for themselves always gets the status `ACTIVATION_REQUIRED` and has to activate it.

What does a fraud report in the PingOne audit log refer to?

Ping documents one place where a user can report fraud. The New Device Paired notification tells the user that a device has been registered to their account, and its template supports a variable that inserts a link for reporting fraudulent pairing attempts. The event list describes `FRAUD.REPORTED` and `DEVICE.FRAUD_REPORTED` only as Fraud Reported. Treat either as the user's statement that they did not pair the device.

Why do the rules for created and deleted devices have a low level?

Every enrollment writes them. A user who gets a new phone creates a device, activates it and deletes the old one. Ping also documents that devices left in the `ACTIVATION_REQUIRED` status for 24 hours are deleted from the system. The activities matter in context, such as an actor who is not the user, an account that was just recovered, or an administrator's account, so the rules are meant for review and for the correlations described under detection logic.

Does a password reset remove an attacker's MFA device?

No. A device stays paired with the user until it is deleted. An attacker who registered a device keeps a valid second factor after the password changes, and with a new password obtained through recovery or phishing can sign on again. Review and clean the list of devices whenever an account is recovered from a compromise.

References

We respect your privacy

We use cookies to keep this site secure and working properly. With your permission, we also use optional cookies to understand usage and improve the experience. Cookie Policy

You can change your choice at any time.