97 Things Every Application Security Professional Should Know

Chapter 90. Will Generative and LLM Solve a 20-Year-Old Problem in Application Security?

Neatsun Ziv

In application security, we find ourselves at a crossroads where the integration of GenAI and LLMs is redefining traditional approaches, offering solutions to the issues of fragmented workflows and excessive tool clutter.

Traditional AppSec models were great at classifying or clustering data based on trained learning of synthetic samples. However, they struggle to keep pace with the hyperactive landscape of techniques, tactics, and procedures used by attackers to exploit any vulnerabilities. Let’s explore the significant impact that GenAI and LLMs can have in transforming the field of application security.

GenAI, with its advanced algorithms and ML capabilities, is proving to be a powerful ally in the battle against vulnerabilities. By analyzing vast volumes of data, including security reports and code samples, GenAI can detect suspicious activities, identify potential malware, and even generate automated fix recommendations with the precision of a seasoned security professional. It is like having an assistant who tirelessly scans and safeguards your applications—and never takes a vacation.

Today’s LLMs are a huge advancement over older models used in ML algorithms that were great at classifying or clustering data based on trained learning of synthetic samples. Modern, sophisticated ...