US-led team introduces framework to make networks of robots, self-driving cars safer

Researchers at Harvard proposed a framework called cy-trust to quantify and manage trust among connected autonomous machines.

The model assigns numeric trust scores (0–1) to information sources and weights their influence on decision-making.

Systems can use onboard sensors and signal-processing techniques to verify data and detect spoofing or impersonation.

Controlled experiments showed cooperative robots reduced trust in suspicious inputs and maintained operation despite simulated attackers.