Sunrun (NASDAQ:RUN) shares rose 3% on Wednesday after the company unveiled a pilot programme designed to turn its network of residential solar and battery customers into a distributed computing platform for artificial intelligence.
The San Francisco-based energy company, which operates around 1.1 million home solar and battery storage systems, has begun testing compute nodes installed in customers’ homes to process AI workloads. The initiative marks Sunrun’s expansion into distributed edge computing, an area the company believes could generate high-margin revenue by leveraging its existing energy network.
The pilot follows a successful proof-of-concept programme that demonstrated both customer demand and the ability to generate revenue from distributed computing capacity. Sunrun is now working to sell AI inference capacity to enterprise customers while evaluating the performance of compute nodes under a range of operating conditions. Homeowners participating in the programme will receive compensation for hosting the equipment.
“AI companies are scrambling to secure greater access to energy and computing power,” said Paul Dickson, Sunrun’s President and Chief Revenue Officer. “Over nearly two decades, we have perfected our ability to operationalize, finance, and scale distributed assets.”
According to the company, demand for AI inference computing is growing by around 35% annually. Citing McKinsey research, Sunrun said inference workloads are expected to overtake AI training by 2030 and account for more than half of global AI computing demand.
Unlike conventional data centres, Sunrun’s model places computing hardware behind the meter in customers’ homes, avoiding the need for land purchases, new transmission infrastructure and lengthy utility interconnection processes. The compute nodes are integrated with the company’s battery storage systems, enabling continued operation during certain grid outages.
Sunrun plans to complete the pilot programme over the coming months before evaluating its performance against key milestones. The company will then decide on the size and pace of a wider rollout and said it is already in discussions with enterprise computing customers, homebuilders and utility partners regarding future commercial deployment.