AI consumes energy in 2 phases: training and inference. The first has a finite cost per training run; the second keeps accumulating with use. One training run has a defined end. Inference follows demand, whether the model handles a few requests or millions. Microblog @antgrasso Training requires computation to develop the model. Once a training run is complete, the energy consumed during that run is a fixed total. The trained model can then be used repeatedly without being trained again for every request. But every request still requires computation. The more frequently the model is used, the more inference energy accumulates. The amount also depends on how much processing each request requires. A simple question and a complex analysis do not necessarily consume the same amount of energy. For organizations deploying AI, estimating energy consumption means considering both the training process and the workload generated throughout the life of the service. #AIEnergy #AIInfrastructure

The full ADIPEC Global Energy Report 2026, Power Shift, is available here: adipec.com/download-adipec-g…

The new ADIPEC Global Energy Report 2026 caught my attention: 22% of oil and gas insiders name AI-driven efficiency as a top commercial opportunity today, versus 27% in ten years. A signal of changing business priorities. Join > adipec.com @ADIPECOfficial Partner. Another finding adds context: 81% of energy insiders expect AI-driven electricity demand to exceed current forecasts. This creates a closer relationship between the energy and technology industries. Energy companies need to understand how computing demand will evolve, and technology companies must understand the capacity and operating requirements of energy systems. I find this mutual dependency particularly interesting because investment decisions in one industry will increasingly affect the other. ADIPEC 2026, taking place from 2-5 November in Abu Dhabi, offers an opportunity to examine how these developments will shape the energy business. Energy. Intelligence. Impact. #ADIPEC2026 #ADIPEC #Energy #AI