AI firmware reverse engineering: devices and hardware taken apart with AI

AI firmware reverse engineering covers the code and signals inside physical devices: microcontroller firmware, appliance images, USB and Bluetooth peripherals, car networks and even FPGA configuration bits. In the projects on this page, a model helped read a firmware dump, decode how a device talks to its app, or map what each bit in a configuration file does. Most of the work ran through Claude Code, which can call command line tools such as OpenOCD, radare2, BlueZ or mpremote while a person handles the hardware. The model reads disassembly and logs quickly, but a human usually has to press buttons, attach probes and confirm that a decoded command really does what the model says. Several entries are research write-ups from security teams rather than finished tools. Each one links the author's writeup or repo and lists the model, tools and reported outcome, so you can see what worked on real hardware.