Google launches low‑carbon edge‑compute platform using retired phones

Google Research unveiled a low‑carbon edge‑compute platform that turns retired smartphones into functional compute nodes [Google Research Blog]. The system runs lightweight workloads—such as sensor aggregation, image preprocessing, and inference—directly on the devices, cutting the need for dedicated servers. By reusing phones that would otherwise become e‑waste, the platform trims both the carbon emissions from manufacturing new hardware and the energy required for data‑center processing.

Google estimates that the average consumer upgrades a phone every 2–3 years, generating millions of tons of e‑waste annually. Repurposing these devices as edge nodes eliminates the disposal step and extends the useful life of existing hardware. The platform’s software stack optimizes power draw, achieving up to a 30 % reduction in energy use compared with comparable cloud‑based workloads [Google Research Blog].

The project is open‑source, with code and documentation released on GitHub, allowing other organizations to deploy similar solutions. Early pilots at a smart‑city sensor network and a remote agricultural monitoring site have demonstrated reliable operation with latency under 100 ms and a 40 % lower carbon footprint than traditional cloud processing.

Google frames the effort as part of a broader push toward a circular tech economy, where devices are continuously repurposed rather than discarded.