Researchers have introduced Qumus, an autonomous AI system designed for experiments with quantum materials. The work has been published on arXiv and is yet to undergo peer review.
This platform integrates robotics, computer vision, and a multi-agent architecture.
Qumus operates within a robotic mini-laboratory, executing the entire process—from protocol planning to result analysis and report preparation.
The system autonomously extracted graphene flakes through mechanical exfoliation and fabricated nanoscale devices from atomically thin materials, including a graphene field-effect transistor based on a van der Waals structure.
To find a method for obtaining a graphene flake larger than 200 µm², the AI explored four parameters: temperature, contact time, the number of pressing cycles, and tape peel-off speed. After over four hours and five optimization cycles, the platform produced a fragment measuring 245 µm².
Qumus is capable of correcting errors. In one test, the system detected the absence of a silicon chip, restructured its plan, and repeated the exfoliation on a new substrate. In another instance, it corrected a misclassification of hexagonal boron nitride as graphene.
During the transistor assembly, the robot performed a 90-minute dry transfer involving 30 physical operations and 18 decision-making stages.
The laboratory is managed by a hierarchical network of agents based on large language models. Computer vision tracks objects and inventory, utilizing YOLOv8 for detection.
In May, Darren Guccione, CEO of Keeper Security, stated that AI and quantum technologies could jeopardize existing security systems.