Executive Summary
Between August 2022 and August 2026, Truffle Security discovered over 9,300 Amazon Web Services (AWS) access keys publicly exposed across code repositories, Git history, datasets, Docker images, and CI logs. Of these, 817 keys were linked to corporate accounts, with 526 being AWS root keys granting unrestricted administrative access. Researchers found that 242 keys belonged to IAM users with AdministratorAccess policies, effectively providing full control over corporate AWS environments. Hugging Face emerged as the largest single source with 8,482 exposed keys, many remaining active for years without rotation.
This incident highlights the persistent challenge of credential management in cloud environments as organizations increasingly rely on Infrastructure as Code and automated deployment pipelines. With the median age of exposed keys being over five years and only 13.7% showing evidence of rotation, the findings underscore critical gaps in security hygiene that threat actors actively exploit for cryptomining operations, data exfiltration, and persistent access establishment.
Why This Matters Now
Cloud credential exposure continues to escalate as organizations accelerate digital transformation and adopt DevOps practices without proper secret management, creating widespread attack surfaces that threat actors increasingly target for initial access and privilege escalation.
Attack Path Analysis
Attackers leveraged exposed AWS access keys found in public repositories, particularly from Hugging Face with 8,482 unique key exposures. With 817 corporate keys and 242 having AdministratorAccess policies, attackers gained immediate privileged access to AWS accounts. They escalated to root-level permissions using the 526 exposed root keys, then moved laterally across cloud services and regions. Command and control was established through legitimate AWS APIs, followed by data exfiltration and potential deployment of cryptominers for sustained financial impact.
Kill Chain Progression
This analysis maps confirmed threat intelligence to the full cloud kill chain to show where defensive gaps would emerge as an attack progresses.
Initial Compromise
Description
Attackers discovered exposed AWS access keys in public repositories, code commits, Docker images, and CI logs, with Hugging Face being the largest source containing 8,482 unique key exposures
MITRE ATT&CK® Techniques
Valid Accounts: Cloud Accounts
Unsecured Credentials: Credentials In Files
Container and Resource Discovery
Valid Accounts: Cloud Accounts
Account Manipulation: Additional Cloud Credentials
Data from Cloud Storage Object
Data Destruction
Resource Hijacking
Potential Compliance Exposure
Mapping incident impact across multiple compliance frameworks.
PCI DSS 4.0 – Strong Cryptography for Authentication Credentials
Control ID: 8.2.1
NYDFS 23 NYCRR 500 – Multi-Factor Authentication
Control ID: 500.15
DORA – ICT Risk Management Framework
Control ID: Article 8
CISA ZTMM 2.0 – Identity and Credential Lifecycle
Control ID: ID.AM-2
NIS2 Directive – Cybersecurity Risk Management Measures
Control ID: Article 21
ISO 27001 – User Registration and De-registration
Control ID: A.9.2.1
Sector Implications
Industry-specific impact of the vulnerabilities, including operational, regulatory, and cloud security risks.
Information Technology/IT
Cloud misconfiguration exposing AWS keys threatens IT infrastructure management, requiring enhanced egress security and zero trust segmentation for multicloud environments.
Financial Services
Leaked AWS root keys enable full account control, risking HIPAA/PCI compliance violations and requiring immediate threat detection for banking infrastructure.
Health Care / Life Sciences
Exposed administrative credentials threaten patient data exfiltration, violating HIPAA 164.312 requirements and necessitating encrypted traffic monitoring and anomaly response.
Computer Software/Engineering
Hugging Face AI platform exposure demonstrates software development risks from credential leakage in code repositories, requiring Kubernetes security and policy enforcement.
Sources
- Hundreds of leaked AWS keys give full control over corporate accountshttps://www.bleepingcomputer.com/news/security/hundreds-of-leaked-aws-keys-give-full-control-over-corporate-accounts/Verified
- Leaked Corporate AWS Keys Held Full Admin Rightshttps://trufflesecurity.com/blog/leaked-corporate-aws-keys-held-full-admin-rightsVerified
- AWS Security Best Practices - Access Keys Managementhttps://docs.aws.amazon.com/general/latest/gr/aws-access-keys-best-practices.htmlVerified
Frequently Asked Questions
Cloud Native Security Fabric Mitigations and ControlsCNSF
Based on the attack progression modeled above, these are the defensive controls that would constrain each stage.
Aviatrix Zero Trust CNSF would have significantly constrained this AWS access key compromise by implementing network segmentation and identity-aware controls that limit lateral movement and reduce blast radius across cloud services and regions.
Control: Cloud Native Security Fabric (CNSF)
Mitigation: Network microsegmentation would likely have limited the scope of services and resources accessible even with compromised credentials, constraining initial foothold establishment across cloud infrastructure components.
Control: Zero Trust Segmentation
Mitigation: Workload isolation policies would likely have constrained the blast radius of administrative privileges, limiting cross-service access even with escalated credentials to reduce account-wide compromise scope.
Control: East-West Traffic Security
Mitigation: Microsegmentation enforcement would likely have blocked unauthorized lateral movement between services and regions, significantly constraining attacker ability to traverse cloud infrastructure boundaries and expand compromise scope.
Control: Multicloud Visibility & Control
Mitigation: Centralized policy enforcement would likely have detected and constrained suspicious cross-region API activity patterns, limiting persistent control establishment across distributed cloud infrastructure and reducing operational persistence.
Control: Egress Security & Policy Enforcement
Mitigation: Controlled egress policies would likely have restricted unauthorized data transfer patterns from sensitive storage services, constraining exfiltration volume and limiting outbound data movement to approved destinations.
Resource-intensive cryptomining operations would likely remain isolated within specific workload segments, limiting computational impact scope and reducing unauthorized charges through constrained resource allocation and network boundaries.
Impact at a Glance
Affected Business Functions
- Cloud Infrastructure Management
- Data Storage and Backup Systems
- Application Hosting Services
- Identity and Access Management
Estimated downtime: N/A
Estimated loss: N/A
Over 9,300 AWS access keys publicly exposed with 817 linked to companies, including 526 root keys and 242 keys with AdministratorAccess policy providing full control over corporate AWS accounts. Potential access to cloud-hosted data, servers, applications, and ability to create persistent admin accounts across affected organizations.
Recommended Actions
Key Takeaways & Next Steps
- • Implement Zero Trust Segmentation with identity-based policies to limit blast radius even when credentials are compromised
- • Deploy Egress Security & Policy Enforcement to prevent unauthorized data exfiltration and cryptominer communications
- • Enable Multicloud Visibility & Control for centralized monitoring of anomalous AWS API usage patterns
- • Configure Threat Detection & Anomaly Response to baseline normal cloud resource consumption and alert on suspicious activities
- • Establish comprehensive credential rotation policies and eliminate root access keys as recommended by AWS security best practices



