The case, explained simply
Google DeepMind · cooling
Help cooling equipment use less energy while keeping a data centre within operating limits.
What actually happened
DeepMind reported up to 40% less energy used for cooling in Google data centres after applying machine learning. The number concerns cooling energy, not the total electricity bill of a data centre. The system used operating measurements to identify more efficient settings.
What the evidence does not tell us
This is a historical, company-reported result in a sophisticated environment. It is not a current product quotation or a promise for Greek hotels, cold stores or factories. An “up to” result is not an average guaranteed saving.
What a Greek organisation can learn
- Ask the facilities manager where energy is used and which settings can safely change. Check meters before buying software.
- Choose one system and compare consumption under similar weather, occupancy or production conditions. Set clear limits that AI cannot change.
- Include comfort, temperature stability, maintenance and staff time in the review. Keep a straightforward way to return to the previous controls.
Before spending money
An energy-efficiency project may need better controls or maintenance before it needs AI. Compare those simpler alternatives in the same budget.
Read the original evidence
Google DeepMind · research account ↗The practical steps above are AIStrategy’s editorial suggestions. They are not reported actions or promised results from the organisation featured. Photography illustrates the subject unless its caption identifies the actual facility.