Unlocking the potential of low consumption AI technologies to accelerate energy transition goals

Artificial intelligence has sparked widespread concern surrounding not only its impact on jobs, but also its voracious appetite for electricity and ultimately its own environmental toll.

Industry leaders speaking at Gastech 2025 in Milan during a panel session titled ‘From investment to impact: Unlocking the potential of low consumption AI technologies to accelerate energy transition goals’ agreed that developing and deploying low-energy, high-impact AI solutions is essential to capitalise on this transformative technology while making the best use of its potential to aid in decarbonisation.

“AI is energy-hungry and it needs to be fueled heavily, and so there's this growing demand for energy to fuel that,” said Uwa Airhiavbere, Chief Commercial Officer for Worldwide Energy & Resources at Microsoft.
Ultimately, by prioritising innovation that minimises energy consumption without compromising impact, stakeholders can ensure artificial intelligence delivers on its decarbonisation promise.

Airhiavbere stated that the true test of AI's potential will not be just its intelligence, but its efficiency, also in terms of power consumption. Microsoft is tech provider as well as a significant energy consumer – he explained- and as such it is aiming to run all data centers in 24 countries on 100% renewable energy by year-end. It has also signed partnerships for nuclear energy and fusion energy to contain its impact on the environment. Forecast models are also part of the strategy.

“We also have an inclusive, 24/7 carbon free matching, where we leverage our AI tools to match renewable energy hour by hour, to make sure that we're not waiting to offset that energy use at the end of the year,” Airhiavbere said.

While artificial intelligence is poised to revolutionise industries and could add an estimated $13 trillion to the global economy by 2030 according to McKinsey, realising that promise depends entirely on sustainably meeting the substantial power demands of its data centers. This puts immense pressure on global energy providers who are already navigating a complex transition.

Long-term sustainable

From the perspective of energy management, the integration of AI is a delicate balancing act, explained Henri Domenach, Global Head of Energy Management at ENGIE.

“Ai is a great opportunity; we need to turn it in a long-term sustainable technology” with a more efficient management of the power needs. Domenach stressed that the integration of AI will itself help optimise the production and reduce carbon footprints. He cited improvements in renewable energy forecasting, crucial for grid stabilisation and value improvement, and a recent commissioning of a 100 MWh battery in Belgium.

Parisa Bardouni, Senior Vice President and Chief Technology Officer at Aker Solutions, said, “There is a lot to gain from collaborations and alliances among technological and energy companies as well as with customers and consumers” to find the best ways to address the AI’s power hunger.

Convergence strategy

“You need to have gas, you need to have nuclear, you need to have renewable — you cannot rely on one specific feedstock because you need to have a resilient grid to power the AI,” said Manoj Narender Madnani, Managing Director for International at MARA.

The convergence of AI and energy management, therefore, presents a clear opportunity to activate transformative solutions. These advancements could serve as a powerful catalyst for a cleaner, more streamlined energy future, driven by innovation in both hardware and software.

Increasingly, “there is a fine line between top tech companies and top energy companies,” Madanani added, saying that he would not be surprised to see some of the former become also energy businesses to better manage the artificial intelligence’s power needs.