Significant security vulnerabilities have been discovered in two of the world’s leading cloud platforms, Microsoft Azure and Google Cloud Platform (GCP), allowing attackers to bypass critical access controls. These “confused deputy” flaws, which can be exploited to acquire administrative level permissions, have left enterprise and government resources at risk.
Justin O’Leary, an independent security researcher, first reported the issues in May and June this year, but neither company acknowledged the vulnerabilities or provided a bug-bounty reward. However, Microsoft has silently patched its flaw. Confused deputy flaws arise when a product or service receives a request from an upstream entity but fails to preserve the original source of the request, allowing it to be forwarded to external parties. This can trick high-privileged entities into accepting requests that appear to come from legitimate sources.
The issue is not new; computer scientist Norm Hardy first coined the term in 1988 after discovering a vulnerability more than a decade earlier in the compiler program at Tymshare, a commercial time-sharing firm in Silicon Valley. Despite significant advancements in identity and access management (IAM) over the past few decades, confused deputy flaws still persist today.
According to O’Leary, many companies have fallen into bad habits when building cloud infrastructures, often using a cookie-cutter approach that prioritizes speed and efficiency over security. “This problem is still around because of these practices,” he says. The vulnerabilities are not isolated incidents but rather a symptom of fundamental weaknesses in modern cloud architectures.
One of the flaws discovered by O’Leary involves Microsoft’s Azure Kubernetes Service (AKS) backup service, which uses a feature called Trusted Access to give the backup vault access to the AKS cluster through specific permissions. However, an attacker can exploit this flaw to escalate privileges from basic Backup Contributor level to achieve cluster-admin privileges for the AKS, allowing them to exfiltrate sensitive data or deploy malicious workloads.
The second flaw involves GCP’s Config Connector, an open source add-on that lets users manage their cloud resources via Kubernetes. According to O’Leary’s research, Config Connector doesn’t perform authorization checks to confirm user identity against Google’s IAM, bypassing access controls. This allows a threat actor equipped with basic Kubernetes namespace access and no GCP permissions to quickly establish themselves as GCP Organization Owner with full administrative control.
The impact of these vulnerabilities is significant; they can be exploited by attackers to gain unfettered access to sensitive data and systems. The issue is further complicated by the fact that malicious actions are often logged as service account activity, disguising the attack.
As a result, it’s essential for cloud users to take proactive steps to mitigate these risks. This includes reviewing their IAM policies, ensuring that all services and products in use preserve the original source of requests, and regularly auditing access controls for weaknesses. By taking these precautions, organizations can minimize the impact of confused deputy flaws and protect themselves against potential attacks.
Source: Dark Reading — 2026-07-27