Could Security Misconfigurations Become No. 1 in OWASP Top 10?

COMMENTARY

The convergence of rising cyber threats, advanced artificial intelligence (AI), remote work, and hybrid infrastructures presents significant cybersecurity challenges in today's IT landscape. As a result, it's necessary to make your endpoints, cloud infrastructure, and remote access channels more secure. As cyber adversaries adopt new tactics, organizations worldwide respond by expanding the use of continuous threat exposure management (CTEM) systems, investing in robust security solutions, and leveraging cross-functional collaboration to mitigate risks and safeguard digital assets effectively.

But like Superman has kryptonite, even the best software has weaknesses, with misconfigurations leading the pack.

Consider this: Microsoft research indicates that a staggering 80% of ransomware attacks can be attributed to common configuration errors in software and devices. 

Misconfigurations now hold an unenviable fifth place on the Open Worldwide Application Security Project Top 10 — a crucial vulnerability reference for the cybersecurity community. OWASP found 208,000 occurrences of common weakness enumeration (CWE) within 90% of applications tested for misconfiguration, highlighting the widespread nature of this vulnerability.

OWASP says, "Without a concerted, repeatable application security configuration process, systems are at a higher risk."

With this evidence, it's no wonder that organizations are paying more attention to "misconfigurations."

Picture This ... 

You're sitting down with your morning cuppa and stories of a data leak hit the headlines. The company affected is a leading insurance firm, and the personal information of thousands of customers has been made available on the Internet for months. With a little research, you learn that the firm left several customer records unprotected on one of its clouds, making it easy for anyone to access this information through a simple SQL command. While digging through the tabloids you stumble upon the cause of such a tremendously ironic turn of events. Turns out, it was a simple misconfiguration error: The system administrator left the cloud open to the public since they missed updating the privacy settings and permissions for the cloud storage in question.

We learn that human errors, despite stringent protocols, are difficult to control and, consequentially, remove. The increasing complexity of distributed and component-based systems and common misunderstandings of system requirements and design will likely lead to more problems. While humans play a critical role in decision-making and monitoring systems, manual updates are no longer viable.

So, What Can You Do About It?

With all that's happening in cybersecurity, can you confidently say you have all your endpoints covered? And by all, I mean all — including the data on third-party systems. If your answer to this is yes, congratulations! You're doing better than most organizations in the world! But if your answer is no, I would like you to consider the following measures to improve the security of your systems: 

While there is a technical aspect to DevSecOps, there is also a human aspect that involves collaboration and planning. A multiprong approach that starts with collaboration across IT operations and security and compliance teams, while discussing the appropriate external and internal compliance requirements, is a critical starting point.

After understanding the configuration and policies, you can start with pre-packaged policies that align with standards such as the Center for Internet Security (CIS) Benchmarks and the Department of Defense Systems Agency-Security Technical Implementation Guides (DISA-STIG). Consider using an automated system to verify if your configurations are continuously accurate. This, in turn, will allow your organization to address complex and heterogeneous environments, including cloud-native public cloud services, Kubernetes configurations, and any on-premises or hybrid cloud workload.