Dr. Kun Dániel Builds an AI-Guided Lockpicking Robot
The maker combines an ESP32 microcontroller with force sensors and Anthropic's Claude to pick pin tumbler locks without human intervention.
By Letizia Giugliano
1 min readHuman-Edited, AI-Assisted
Maker Dr. Kun Dániel has designed an autonomous lockpicking device that replaces the human touch with robotic actuators and artificial intelligence. The compact invention mounts onto an ordinary lock cylinder to manipulate internal pins through real-time physical feedback.
Sensors and cloud computing replace human touch
Traditional lock manipulation relies on fine motor control and tactile sensitivity to feel when internal pins reach the shear line. Instead of relying on manual dexterity or simple brute force, the automated tool created by Dr. Kun Dániel copies that sensory feedback loop electronically, as reported by Hackster.io News.
At the center of the hardware sits an ESP32-C3 Mini development board. The microcontroller monitors forces inside the pin tumbler lock and connects via Wi-Fi to Claude, an artificial intelligence model developed by Anthropic. By offloading decision-making to the model, the machine evaluates mechanical resistance pin by pin to decide its next physical action.
Mechanical assembly built from custom 3D prints
To translate algorithmic commands into physical action, the build relies heavily on custom mechanics and 3D printing. The moving assembly integrates one stepper motor, two servo motors, and two load cells configured to gauge minute resistance levels.
According to Hackster.io News, a 3D-printed rack-and-pinion setup converts rotational motor drive into precise linear motion for the picking probe. The unit clamps straight onto the cylinder body, draws electricity from a rechargeable lithium-ion battery, and reports pin measurements live on an integrated OLED display.
Open-source plans following testing and refinement
The creator has already invested more than 200 hours into refining the hardware and software. Video demonstrations show the functional prototype rotating a cylinder successfully, though software optimization remains underway to boost overall reliability.
Once testing concludes, Dániel plans to publish the firmware source code, documentation, and 3D design files publicly so other builders can study the system, according to Hackster.io News.
Read also on The Patent: Trilateral patent offices meet in Vienna on Global Assignment and AI
