Initial Seeds Generation Using LLM for IoT Device Fuzzing

In recent years, security measures for IoT devices have become more important. Fuzzing of IoT devices is an effective way to find unknown vulnerabilities. In IoT device fuzzing, a large number of test cases are generated based on a set of initial seeds and sent to a target device to monitor its behavior, in which a device crash means that a vulnerability has been discovered. However, generating a set of initial seeds is difficult because technical knowledge in security and adaptation to various IoT devices are quite required. In this paper, we propose a method to generate initial seeds for IoT device fuzzing effectively utilizing a large language model (LLM). The proposed method generates initial seeds for efficient fuzzing of the target IoT device by inputting only the specifications of the IoT device and the name of the vulnerability to be inspected into an LLM, with no specific technical knowledge in security. Experimental results show that the proposed method detected crashes in 5.4 minutes and 12.43 minutes from the start of fuzzing, that could not be detected by the initial seeds generated manually.