AWS CVE-2026-5708

Requires authenticated access to AWS Research and Engineering Studio (RES) versions prior to 2026.03. Attacker must be an authenticated remote user with valid credentials to craft malicious API requests targeting the session creation component to escalate privileges and assume virtual desktop host instance profile permissions.

This vulnerability presents significant real-world risk in AWS environments. The CVSS 4.0 score of 8.7 reflects high impact across confidentiality, integrity, and availability (VC:H/VI:H/VA:H) with network-based attack vector (AV:N), low attack complexity (AC:L), and requiring only low privileges (PR:L). Authenticated attackers with basic RES access can escalate to cloud infrastructure permissions, potentially accessing sensitive research data, modifying cloud resources, or pivoting to other AWS services. The attack is network-exploitable and does not require user interaction (UI:N), making it suitable for automated exploitation once initial access is obtained. While no public exploit identified at time of analysis and EPSS probability data not provided, the vulnerability's technical characteristics (mass assignment, cloud privilege escalation, low complexity) make it an attractive target. The presence of a vendor-released patch (2026.03) and public GitHub issue discussion increases risk of reverse-engineering. Organizations running RES in production AWS environments should prioritize remediation given the potential for lateral movement and cloud resource compromise. The requirement for authentication (PR:L) provides some defense-in-depth but does not significantly reduce risk in environments with broad RES user access.

An attacker with valid RES user credentials (low privilege requirement) authenticates to the RES platform and crafts a malicious API request to the session creation endpoint. By injecting unsanitized attributes into the session creation parameters, the attacker manipulates object properties to assume the IAM instance profile attached to virtual desktop EC2 hosts. With escalated cloud permissions, the attacker queries S3 buckets containing research data, modifies security groups to expose internal resources, or launches additional compute instances for cryptomining. The low attack complexity and network-based vector enable remote exploitation without physical access or user interaction, making this suitable for targeted attacks against research environments or supply chain compromise scenarios.