How Service Providers Can Capitalize on the Growth of Edge AI

The cloud has significant backhaul costs and cannot support real-time, mission-critical AI applications. Many AI applications such as self-driving cars and video surveillance require near-real-time data to work effectively. Exposing proprietary data to the cloud is also a concern. Leaking personal information or proprietary intellectual property is not an option—security must be top of mind when designing an Edge AI architecture.

Service provider networks are the first point of connectivity for AI-powered devices. This means service providers are perfectly positioned to be the key insertion point for Edge AI resource utilization—if they can provide the infrastructure to support their customers. Service providers that can deliver an application service (AS) based on Edge AI technologies will be able to monetize delivery of key services to customers with AI-driven devices.  

Edge AI investments are rapidly accelerating, with annual network Edge AI spend expected to increase to more than $4 billion by 2030. To effectively support the AI applications of their customers—and generate healthy over-the-top AI-related revenue—service providers must start investing in their Edge AI infrastructures today. 

Service providers are poised to generate over $2.5 billion annually in revenue from AI applications. Current AI architectures do not leverage Edge AI infrastructure, but businesses are willing—and in many cases required—to leverage Edge AI solutions to optimize their AI applications. Service providers will need to build and offer Edge AI infrastructure to their customers to provide the Edge AI capabilities that businesses need. This will translate into significant increases in the utilization of Edge AI resources.

Edge AI is transforming the transport sector. Connected car driver assistance is one of the most significant use cases for Edge AI workloads—low latency and local inference are critical to the success of the AI application, preventing accidents and optimizing driving patterns. The transport sector is expected to account for more than 60% of network edge revenue across the forecasted period.

Smart cities utilize real-time traffic monitoring, video surveillance, and resource management to optimize their infrastructure and security. By 2030, 54% of smart city AI workloads will run at the network edge, moving away from on-device and on-premises processing. Improved latency and localized information will enhance their ability to operate efficiently and securely.