There’s an ongoing struggle for balance between the usefulness of AI and its energy demands. Generative AI and large language models are highly energy-intensive, with ChatGPT consuming up to 10 times more electricity than a standard Google search, and the energy consumption of data centres projected to double by 2026.

But at the same time, AI has the potential to significantly improve energy efficiency. Co-host Laura Young Laura speaks with Ben Sigrin, Senior Product Manager at GridBeyond, about how the company uses AI to manage energy behind-the-meter (on the consumer side), optimising energy use in buildings and reducing carbon emissions.

AI is also transforming sectors like transportation and agriculture by enhancing overall energy efficiency. Digital twins - virtual models that replicate real-world conditions—could eliminate 5-10% of global greenhouse gas emissions by 2030. NVIDIA, which you my know as the maker of the graphic processors that are used to train AI, has built its own version of these digital twin models. Director of Product Marketing Shar Narasimhan talks to Carolyn about Earth2, a mammoth digital replica of Earth used to model the extreme weather and climate change that affect our energy networks, and about how AI energy demands are being addressed in the industry.