• • • 2026 PREDICTIONS • • •
2026 PREDICTIONS: EVOLVING DATA CENTRES FOR AN AI-DRIVEN FUTURE
AI’s early disruptions were only the warm up. By 2026, the rubber will hit the road and AI will become ever more ingrained in every aspect of life as the focus shifts from Large Language Models (LLMs) to AI inferencing.
By Steven Carlini, Vice President of Innovation and Data Centre, Schneider Electric
rganisations across every sector are accelerating AI adoption with it transforming fields ranging from academia to life sciences, finance and manufacturing, and even data-centre management itself. This surge is driving denser AI workloads which is creating accelerated demand for advanced cooling, increased retrofitting of existing facilities, greater focus on building successful AI factories and expanded use of digital twins for greater efficiency. With geopolitical uncertainty and rapid technological change, resilience and adaptability will remain critical for data centres. Let’s look at what we can expect in 2026 and beyond.
O
AI transforms functions and companies
AI is playing a role in many business functions and its use is increasing. According to McKinsey’s annual survey on the state of AI, 78 per cent of organisations use AI in at least one business function, compared to 72 per cent in early 2024 and 55 per cent the year before. It’s most often used in sales and marketing, according to the McKinsey survey, but its use will expand into other arenas, including manufacturing and supply chains, medical facilities, financial institutions and data centres.
• Manufacturers using AI to support demand forecasting, a longstanding challenge in the industry, are able to improve their forecast accuracy by a median 30 percentage point.
• Hospitals increasingly use predictive AI to simplify or automate billing procedures, help with appointment scheduling and identify high-risk outpatients for follow-up care.
• Financial institutions are using AI to enhance fraud detection, payment optimisation and risk management for businesses.
• In data centres, AI-driven cooling systems use predictive analytics to minimise overheating and energy waste. AI can also enhance grid efficiency, integrate renewable energy sources, and lower carbon emissions by analysing vast amounts of data to predict and balance electricity supply and demand.
As adoption continues to grow, businesses will not only rely on AI, but see it transform their companies and industries themselves. AI agents operating with little or no supervision will become core to business processes. These agents will rely on various models working together and require significant processing power and data centre capacity or an AI factory.
The rise of AI factories An ‘AI factory’ is essentially a data centre that outputs intelligence, rather than simply storing and processing data. We are moving beyond AI training models to inferencing, which is where businesses will see ROI from AI. An AI factory trains a model, fine tunes it and then performs inference, generating valuable intelligence that can be sold or used to gain a competitive advantage.
While typically requiring less power per server than training, inferencing workloads are increasingly varied and pervasive. They now range from simple chatbot prompts to complex real-time analysis in healthcare, retail and other industries using autonomous systems and agentic agents. Depending on the deployment and workload, inference environments can range from less than 20kW for compressed or tuned models, up to 140kW per rack for more advanced agentic use cases.
By 2030, we expect the data centre market to reflect this diversity:
• 25 per cent of new builds will be over <40kW per rack, primarily inference focused.
• 50 per cent will fall into the 40–80kW per rack range for mixed inference and training workloads.
36 ELECTRICAL ENGINEERING • DECEMBER/JANUARY 2026
electricalengineeringmagazine.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52
