Google Research has introduced SensorFM, a project described as a step toward a general intelligence and interface for wearable health data. The work is framed under the company's generative AI efforts, suggesting that the same underlying approach used for language and multimodal models is being applied to physiological sensor streams.
The stated ambition is to move beyond task-specific wearable analytics and toward a foundation model that can understand and work with diverse sensor data. Rather than treating each health application as a separate modeling problem, SensorFM appears designed to serve as a reusable layer for many downstream tasks involving signals collected from wearable devices.
Because this article is based on a single source—Google Research's blog post—there are no differing sources to compare. The claims above reflect Google Research's own framing of the project and its goals.