Attack scenario · AWS identity & access

AWS IAM Identity Center permission misuse

How can IAM Identity Center permission sets be misused? An identity able to edit permission sets or account assignments grants itself administrator access across many accounts at once.

Fictional, educational scenarioPrivilege takeoverLast updated · ATT&CK v20.1

Attack path · 3 stages

  1. 1

    Control an identity with Identity Center admin rights

    T1078.004

  2. 2

    Modify a permission set or assignment

    T1098.003T1548.005

  3. 3

    Use the new access across accounts

    T1021.007

For leadership

Executive summary

IAM Identity Center is the recommended way to give people access to many AWS accounts. That centralization is valuable — and means the ability to edit permission sets or assignments is equivalent to administrator access across every account they reach. If that ability sits with a broad group, or with a help-desk tool, one compromised identity can quietly grant itself organization-wide privileges. Leaders should treat permission-set administration as one of the most sensitive roles in the company. That means a small named group, strong authentication, an approval step for changes to administrative permission sets, and an alert every time an assignment changes — so that a new administrator in production is always a deliberate, visible decision rather than a quiet side effect of an access request.

For engineers

Technical summary

Principals with sso:CreateAccountAssignment, sso:PutInlinePolicyToPermissionSet, sso:AttachManagedPolicyToPermissionSet or sso:ProvisionPermissionSet (in the management or delegated-administrator account) can grant themselves or their group elevated access in member accounts. Changes to the identity source (group membership) have a similar effect.

Environment and assumptions

Fictional company: Identity Center administration is delegated to a "platform-tools" account. A support automation role there has sso:* to fix access requests quickly. The role’s credentials are available to a ticketing integration.

  • The adversary controls the support automation role or a user in the admin group.
  • Changes to permission sets are not alerted on.
  • Fictional and simplified for education.

Interactive

Attack sequence

Select a stage, or use the arrow keys, to see preconditions, evidence and controls.

Stage 1

Control an identity with Identity Center admin rights

The adversary uses the support automation role.

Preconditions
  • Role has sso:* in the delegated admin account
Required access
  • Automation role session.
Potential evidence
  • CloudTrail in the delegated admin account
Defensive controls
  • Limit Identity Center admin rights to a small, MFA-protected group
Text version of the attack sequence
  1. Control an identity with Identity Center admin rights (T1078.004). The adversary uses the support automation role. Preconditions: Role has sso:* in the delegated admin account. Evidence: CloudTrail in the delegated admin account. Controls: Limit Identity Center admin rights to a small, MFA-protected group.
  2. Modify a permission set or assignment (T1098.003, T1548.005). The adversary attaches AdministratorAccess to a permission set they are assigned, or assigns themselves an admin permission set in production accounts. Preconditions: sso:CreateAccountAssignment / AttachManagedPolicyToPermissionSet. Evidence: CloudTrail: CreateAccountAssignment, AttachManagedPolicyToPermissionSet, ProvisionPermissionSet. Controls: Alert on all permission-set and assignment changes; Change approval for admin permission sets.
  3. Use the new access across accounts (T1021.007). The adversary signs in to production accounts with the elevated permission set. Preconditions: Assignment provisioned. Evidence: CloudTrail in member accounts: AssumeRoleWithSAML into AWSReservedSSO_* roles. Controls: Short session durations for admin permission sets; Periodic assignment review.

Technique mapping

MITRE ATT&CK and ATLAS techniques referenced in this scenario
IDTechniqueTactics (official)Stages
T1078.004Valid Accounts: Cloud AccountsInitial Access, Persistence, Privilege Escalation, Stealth1
T1098.003Account Manipulation: Additional Cloud RolesPersistence, Privilege Escalation2
T1548.005Abuse Elevation Control Mechanism: Temporary Elevated Cloud AccessPrivilege Escalation2
T1021.007Remote Services: Cloud ServicesLateral Movement3

Technique names and tactic assignments come from the imported MITRE data. Mapping a stage to a technique is Securitain’s analysis.

Why the attack can succeed

  • Permission-set administration is treated as an operational task rather than a privileged one.
  • One change propagates to many accounts.
  • Automation was given sso:* to avoid friction.

What prevents or limits the attack

What may detect the attack

Detection signals and telemetry sources
SignalTelemetry sourceNotes
Permission set policy changes and new account assignmentsCloudTrail (management / delegated admin account)—
Group membership changes in the identity sourceIdP audit logs—

Telemetry only helps if it is enabled before the incident and someone acts on it. Listing a signal here does not mean it is collected or monitored in your environment.

What Securitain may assess

  • Securitain can assess todayInventory of IAM users, roles, groups, policies, access-key configuration and trust relationships across connected AWS accounts. Learn more
  • Securitain can assess todayEffective-permission view that considers managed, inline, group, boundary and SCP policies together. Learn more
  • Outside Securitain’s current scopeReal-time detection of attacker activity in CloudTrail is not part of Securitain’s current assessment — use GuardDuty, a SIEM or your SOC for that.

For leadership

Executive implications

integrity
Administrator access across many accounts at once.
compliance
Access-review controls are undermined if assignments change without approval.

No financial loss estimates are given; impact depends on the data, contracts and regulation involved.

Next steps

  1. List who can modify permission sets and assignments.
  2. Remove sso:* from automation.
  3. Alert on every assignment change.

References

MITRE ATT&CK® and ATT&CK® are registered trademarks of The MITRE Corporation. ATT&CK data © 2026 The MITRE Corporation. This work is reproduced and distributed with the permission of The MITRE Corporation (ATT&CK v20.1). Securitain is not affiliated with or endorsed by MITRE. Explanations, scenarios and control mappings on this page are Securitain’s own analysis. Sources and licensing

Stay in the loop

Cloud security insights, early access & compliance updates

Weekly InsightsEarly AccessExpert Content
No spam, everUnsubscribe anytimeWeekly updates
Chat with Securitain