TA0003Tactic 5 of 15

Persistencein cloud and AI environments

How do attackers maintain access to an AWS account?

ATT&CK v20.1Updated Official tactic page
Persistence technique map22 techniques in Persistence, 11 of them apply to cloud platforms. Highlighted: T1098.001, T1098.003, T1136.003, T1525, T1078.004.T1078T1098T1136T1525TA000322 TECHNIQUES
Key for cloud (this guide)Applies to cloud platformsOther platforms
22
Techniques
91
Sub-techniques
11
Apply to cloud platforms
4
Defensive controls
3
Detection signals

What the tactic means

Persistence covers techniques adversaries use to keep access across credential rotations, restarts, redeployments and partial remediation. It is why an incident is not over when the first compromised credential is disabled.

The adversary is trying to maintain their foothold.

Persistence consists of techniques that adversaries use to keep access to systems across restarts, changed credentials, and other interruptions that could cut off their access. Techniques used for persistence include any access, action, or configuration changes that let them maintain their foothold on systems, such as replacing or hijacking legitimate code or adding startup code.

MITRE ATT&CK TA0003 description

Why it matters: cloud, AWS and AI

Cloud environments

Cloud persistence is mostly identity persistence: additional credentials, new principals and modified trust. It survives because responders revoke what they know about and miss the rest.

AWS specifics

Creating access keys or login profiles for other users, creating roles trusting external accounts, attaching policies, adding OIDC/SAML providers, implanting images in ECR or AMIs, and scheduling functions are typical AWS persistence methods.

AI & workloads

AI systems add persistence surfaces: poisoned knowledge-base content, modified agent configuration or memory, and malicious tools installed into agent environments.

AWS API calls to know
  • iam:CreateAccessKey / CreateLoginProfileAdd credentials to existing users.
  • iam:CreateRole / UpdateAssumeRolePolicyCreate or modify trust for external access.
  • iam:CreateOpenIDConnectProvider / CreateSAMLProviderAdd an identity provider the adversary controls.

What the attacker is trying to achieve

  1. 01Survive credential rotation
  2. 02Create hidden administrative paths
  3. 03Implant code in images or automation

Techniques that matter most in cloud

What it needs, and what breaks it

Attacker needs

  • IAM write permissions
  • Unmonitored identity changes

Defenders control

  • Restrict IAM write to a small set of principals
  • Permission boundaries for delegated role creation
  • SCPs on credential creation and external trust
  • Incident runbook that enumerates all identity changes and revokes sessions

Detection signals and telemetry

Signal
  • CreateAccessKey, CreateUser, CreateRole, CreateLoginProfile

    CloudTrail
  • Trust policies granting unknown external principals

    IAM Access Analyzer
  • New identity providers

    CloudTrail

For leadership

What this means for the business

An incident is only over when every foothold is removed. Ask responders for evidence that all identity changes during the intrusion window were reviewed — not just that the leaked key was rotated.

Common misconceptions

Myth

“Rotating the leaked key ends the incident.”

Reality

Additional credentials, roles and active sessions remain unless explicitly found and revoked.

Practical security checklist

0 of 4 done

Saved only in this browser. Nothing is sent to Securitain.

Related attack scenarios

Where Securitain fits

  • 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 todayRisky trust relationship analysis, including roles trusted by external accounts and missing or unscoped ExternalId. Learn more
  • Securitain can assess todayPrivilege-escalation risk analysis for supported combinations such as iam:PassRole with Lambda, EC2, ECS or CloudFormation. Learn more

Official sources and 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