Leaked AWS Keys with Admin Access Found Active by Truffle Security

Thousands of Leaked AWS Keys, Many with Full Admin, Remain Active

HIGH
August 23, 2026
5m read
Cloud SecurityData BreachSecurity Operations

Related Entities

Organizations

Products & Tech

Full Report

Executive Summary

A new report from Truffle Security has exposed a critical and widespread failure in cloud credential management. Their investigation found that thousands of Amazon Web Services (AWS) access keys, leaked publicly over the past four years, are still active. Alarmingly, 768 of these active keys provide full administrative privileges, including 526 highly-sensitive root access keys and 242 IAM user keys with the AdministratorAccess policy. These credentials, found in public sources like Git histories and Hugging Face datasets, grant attackers the ability to completely take over corporate cloud accounts. The findings highlight a systemic issue with secret hygiene and the inadequacy of relying solely on automated detection and quarantine measures.

Threat Overview

The research, conducted between August 2022 and August 2026, involved collecting 431,875 AWS secrets from various public sources. After deduplication, 64,024 unique keys were identified. The security firm then tested a sample of 10,616 key pairs and discovered that 88% were still active as of August 10, 2026. The median age of the exposed keys was approximately five years, indicating a long-term failure in credential rotation and revocation processes.

Hugging Face, a popular platform for AI models and datasets, was identified as the single largest source of leaks, with 8,482 unique keys discovered. This suggests that developers are inadvertently committing secrets within their machine learning projects.

Technical Analysis

The most severe finding is the number of active root keys (526). A root key is tied to the account owner and has unrestricted access to all resources within the AWS account. Compromise of a root key is equivalent to a complete account takeover. The 242 active IAM keys with AdministratorAccess are similarly dangerous, allowing full control over IAM policies and all other AWS services.

MITRE ATT&CK Techniques:

The research also noted that AWS's automated quarantine policy for leaked keys, AWSCompromisedKeyQuarantineV2, may not be sufficient. While it restricts many actions, it can still allow an attacker to assume other roles, stop audit logging, or execute commands on existing EC2 instances, providing a foothold for further escalation.

Impact Assessment

The impact of a compromised administrative AWS key is catastrophic. Attackers can:

  • Steal Sensitive Data: Exfiltrate entire databases, customer information, and intellectual property from S3 buckets, RDS, and other services.
  • Hijack Resources: Deploy powerful EC2 instances for cryptocurrency mining, costing the victim organization hundreds of thousands of dollars.
  • Deploy Ransomware: Encrypt critical data and demand a ransom for its release.
  • Disrupt Operations: Shut down or delete production infrastructure, causing major business outages.
  • Establish Persistent Access: Create new IAM users and backdoors to maintain long-term access to the environment.

Cyber Observables — Hunting Hints

Organizations should proactively hunt for exposed secrets. The following patterns can help identify potential leaks:

Type
String Pattern
Value
AKIA[0-9A-Z]{16}
Description
The standard prefix for an AWS Access Key ID. Scanning code repositories for this pattern is a common first step.
Type
File Path
Value
~/.aws/credentials
Description
The default location for storing AWS keys on a developer's machine. Its presence in a public repository is a critical finding.
Type
Log Source
Value
CloudTrail Logs
Description
Monitor for API calls originating from unusual geographic locations, IP ranges (e.g., Tor exit nodes), or using keys that have been inactive for a long time.

Detection & Response

  1. Secret Scanning: Implement automated secret scanning tools (like TruffleHog, GitGuardian, or GitHub Advanced Security) in your CI/CD pipeline to prevent secrets from ever being committed to code repositories.
  2. CloudTrail Analysis: Actively monitor AWS CloudTrail logs for suspicious API activity. Look for high-risk API calls like StopLogging, CreateUser, or PutRolePolicy from unexpected sources.
  3. Third-Party Monitoring: Utilize services that scan public repositories for secrets associated with your organization.

Mitigation

  • Eliminate Root Key Usage: The AWS root user access key should never be used for programmatic access. It should be deleted, and MFA should be enabled on the root account. This is a critical best practice.
  • Principle of Least Privilege: Grant IAM users and roles only the permissions they absolutely need to perform their tasks. Avoid using wildcard (*) permissions.
  • Use IAM Roles for EC2: Instead of embedding long-lived access keys in applications, use IAM Roles for EC2 instances, which provide temporary, automatically rotated credentials.
  • Regular Key Rotation: Implement and enforce a strict policy for rotating all IAM user access keys (e.g., every 90 days).
  • Developer Training: Train developers on the risks of hardcoding secrets and provide them with secure alternatives like AWS Secrets Manager or HashiCorp Vault.

Timeline of Events

1
August 10, 2026
Truffle Security conducted tests on a sample of leaked AWS keys, finding 88% were still active.
2
August 23, 2026
This article was published

MITRE ATT&CK Mitigations

Utilize secret management systems and avoid hardcoding credentials in code or configuration files.

Strictly limit the use of highly privileged and root accounts. Use IAM roles and temporary credentials instead of long-lived keys.

Enforce regular rotation of all access keys and promptly revoke any keys that are no longer needed or are suspected of compromise.

Enforce MFA on all accounts, especially the root account, to provide an additional layer of security against credential compromise.

Timeline of Events

1
August 10, 2026

Truffle Security conducted tests on a sample of leaked AWS keys, finding 88% were still active.

Article Author

Jason Gomes

Jason Gomes

• Cybersecurity Practitioner

Cybersecurity professional with over 10 years of specialized experience in security operations, threat intelligence, incident response, and security automation. Expertise spans SOAR/XSOAR orchestration, threat intelligence platforms, SIEM/UEBA analytics, and building cyber fusion centers. Background includes technical enablement, solution architecture for enterprise and government clients, and implementing security automation workflows across IR, TIP, and SOC use cases.

Threat Intelligence & AnalysisSecurity Orchestration (SOAR/XSOAR)Incident Response & Digital ForensicsSecurity Operations Center (SOC)SIEM & Security AnalyticsCyber Fusion & Threat SharingSecurity Automation & IntegrationManaged Detection & Response (MDR)

Editorial Standards & Analyst Review

CyberNetSec.io uses automation to assist source monitoring, deduplication, observable extraction, and structured intelligence generation. Published analysis follows human-defined editorial standards and adds defensive context including MITRE ATT&CK, D3FEND, STIX, and Sigma where applicable. Read our editorial policy.

Tags

AWSCloud SecurityAccess KeysIAMSecret ScanningData Leak

📢 Share This Article

Help others stay informed about cybersecurity threats

🎯 MITRE ATT&CK Mapped

Every tactic, technique, and sub-technique used in this threat has been identified and mapped to the MITRE ATT&CK framework for consistent, actionable threat language.

🧠 Enriched & Analyzed

Observables and indicators of compromise (IOCs) have been extracted and cataloged. Risk has been assessed and correlated with known threat actors and historical campaigns.

🛡️ Actionable Guidance

Detection rules, incident response steps, and D3FEND-aligned mitigation strategies are included so your team can act on this intelligence immediately.

🔗 STIX Visualizer

Structured threat data is packaged as a STIX 2.1 bundle and can be visualized as an interactive graph — relationships between actors, malware, techniques, and indicators.

Sigma Generator

Sigma detection rules are derived from the threat techniques in this article and can be converted for deployment across any major SIEM or EDR platform.