Analysts expect AI inference to keep moving out of data centers and onto edge devices, gateways and on-premises nodes, and some think enterprise adoption is close to a turning point. "I think lots of vendors and different parts of the ecosystem I've spoken with are looking at 2026, 2027 for there to be an inflection point," Paul Schell, an analyst at ABI Research, told Network World.

The forecasts point the same way. Gartner predicts that by 2028 more than two-thirds of enterprise-managed data will be created and processed outside the data center or cloud, and that more than two-thirds of enterprises will deploy edge AI by 2029, up from 10% in 2025. IDC expects half of all enterprise AI inference workloads to run on endpoints or edge nodes by 2030. The EdgeAI Foundation puts edge AI's growth rate at 37% a year through 2030, compared with 28% for the AI market overall.

Analysts See Edge AI Inference Nearing an Enterprise Inflection Point

Why inference is moving

Volume is one reason. Gartner counted about 11.7 billion installed IoT devices in 2025, growing 9% a year, and analyst Thomas Bittman says as much as 90% of edge data goes unprocessed today. Cost and latency are others. "AI is moving to the edge because latency, cost, and resilience demand it," said Olga Yashkova, IDC's research manager for edge AI strategies. Industrial automation and video analytics need onsite inference, she said, because sending data to the cloud and back "is too slow, too expensive and too risky when connectivity is unreliable."

Data control matters too, especially in Europe, where residency and sovereignty rules are strictest, Schell said. Sensor data increasingly includes biometrics such as facial images collected for safety or security.