From MEC to edge AI: capturing the distributed AI infrastructure opportunity

Operators are increasingly launching GPU as a service (GPUaaS) and sovereign AI services in centralised data centres to tap into AI demand. However, their ability to capture market share will be limited. Edge AI can extend these deployments across operators’ distributed network assets, opening up a broader set of market opportunities.

Information included in this report

• An overview of limitations in telecoms operators’ market opportunity in centralised data centre-based AI infrastructure offerings
• An exploration of the emerging edge AI opportunity with the shifting AI workload demands from training to inference
• An assessment of how edge AI differs from multi-access edge computing (MEC) and how this creates new opportunities for telecoms operators
• An outline of edge AI infrastructure layers and where telecoms operators are better positioned to play
• A framework for identifying the right edge AI use cases where telecoms operators have structural advantages