Comcast Corporation (NASDAQ: CMCSA) announced a field trial to deploy NVIDIA GPUs at the edge of its network infrastructure to test AI applications closer to customers. The collaboration aims to reduce latency for AI workloads by processing them in regional facilities rather than distant data centers.

The trial leverages Comcast's distributed network architecture that reaches 65 million homes and businesses. The company will test three specific applications: a personalized advertising system using Decart's real-time AI video models, a small business concierge service powered by Personal AI's technology on HPE ProLiant servers, and reduced latency gaming services.

The personalized advertising system will customize video advertisements at the household level using factors such as language preferences and household demographics. The small business service will provide AI-powered customer greeting, appointment management and operational support. The gaming application builds on Comcast's existing low-latency technology for NVIDIA GeForce NOW.

"The industry is shifting towards a more distributed AI infrastructure and Comcast operates a network that supports it today," said Elad Nafshi, Chief Network Officer at Comcast. The company's network includes DOCSIS 4.0 FDX nodes, smart amplifiers and intelligent gateways designed to support real-time AI inference.

"Distributed AI Grid is the next big opportunity for the telecommunications industry, and Comcast's nationwide, deeply distributed network is a perfect match for building it," said Ronnie Vasishta, SVP of AI and Telecoms at NVIDIA.

The companies plan to explore additional opportunities including AI-enhanced advertising, small business solutions, premium gaming services and third-party edge compute services. Comcast will provide updates on the initiative at GTC in San Jose.