The answer to AI’s energy needs may be blowing in the wind

Data centers and offshore wind farms could prove to be a perfect match as AI becomes increasingly energy-hungry.

Many of the companies racing to implement new generative AI tools have also made major commitments to controlling their greenhouse gas emissions. A wave of new AI data centers could undermine these climate goals – unless they find a clean source of energy.

Conveniently, some of the peculiarities of AI data centers make them a good match for offshore wind farms. And the offshore wind industry, which has already joined forces with big tech companies, could certainly use a boost as it forecasts more uncertain economic waters.

“Data centers and AI… We are excited about this opportunity,” said Mads Nipper, CEO of leading offshore wind developer Ørsted, in a statement. earnings call last week. “We are excited about this because it is very high growth, in demand, but also because [comes from] companies that are generally very committed to transforming their [carbon] footprint.”

Giants like Microsoft and Google are investing everything they have into developing the best AI tools and putting them into every conceivable platform, from games to social media to work apps and search. All of this leads to high energy costs.

By 2026, AI could consume 10 times more electricity than in 2023, according to the International Energy Agency. The data centers used to train AI consume even more energy than the traditional data centers used to monitor emails and cat videos from around the world. And AI data centers are taking up more space than ever before. Data centers under construction in North America combined would need as much energy as all the homes in the San Francisco metro area annually, according to a February report. report from real estate services company JLL.

“If we could make these data centers close to [offshore wind farms], this is something that can offer high loads of attractive power. Then [it’s] something we will certainly dive into in Europe and the US to take advantage of this opportunity,” Nipper said on the earnings call.

Flexibility is what can make data centers for AI particularly attractive to offshore wind developers. Unlike other data centers, those used to train AI models don’t have to worry as much about being close to population centers to reduce latency. They can also theoretically ramp up and shut down operations more easily to match fluctuations in the weather, which can be challenging when relying on wind power. This means AI data centers could harness surplus energy that might otherwise go to waste.

“This is really unique because if you look at the big picture, not many industries are flexible [with their energy use],” says Odin Foldvik Eikeland, an analyst at research firm Rystad Energy. “It could also open up some really cool collaborations… I don’t see any reason why not.”

For technology companies, offshore wind farms are not only attractive as an abundant source of carbon pollution-free energy. There are other potential benefits. Microsoft has launched a research project exploring the feasibility of subsea data centers powered by offshore wind farms called Natick Project. On land, data centers have sparked concerns about how much electricity It is water they use for refrigeration systems. Storing servers under the sea could be a more sustainable way to keep them cool. Keeping servers submerged can also prevent corrosion caused by oxygen and moisture. After two years, Microsoft’s Project Natick servers had only one-eighth the failure rate of terrestrial servers.

Tech companies have made a wave of climate commitments in recent years, becoming the largest corporate buyers of renewable energy. Microsoft, which has invested more than $13 billion in OpenAI, last week made the largest corporate deal to purchase renewable energy to date. Google, in its “Gemini era”, announced in February its largest offshore wind deal to power data centers in Europe. Both companies have committed to matching 100% of their electricity consumption with carbon-free energy purchases, 24 hours a day, by 2030.

This has been a boost for offshore wind developers like Ørsted at a time when they face strong economic headwinds. The offshore wind industry is still trying to take off, accounting for less than 1 percent of the global electrical mix. And supply chain disruptions and rising inflation rates have paralyzed the industry since the Covid-19 pandemic, leading Ørsted to cancel two major projects in the USA at the end of last year.

Even so, public services are running out of time to figure out how to meet the growing demand for electricity. The electrification of homes, buildings and vehicles puts more pressure on electrical grids. The boom in new data centers used for crypto mining and AI is another major challenge. And everything is happening at the same time.

“For starters, data center computing needs a lot of power, and then AI is just a magnitude higher than that,” says Stephen Maldonado, research analyst for North American wind energy at Wood Mackenzie. “I think it’s a real possibility that we could see this type of demand drive utilities to look to offshore wind as a way to achieve these decarbonization goals.” Offshore wind farms are especially attractive in densely populated coastal regions, he says, where there isn’t much space on land to erect skyscraper-sized turbines.

Whatever happens next will have to be a balancing act. The US, for example, has not yet installed enough renewable energy to meet its climate goals. And the emergence of new AI data centers risks diverting clean energy from arguably more important things, like keeping the lights on in people’s homes. But if data centers don’t run on clean energy, fossil fuels will be the alternative. Pipeline operator TC Energy Corp said on an earnings call last week which expects new data centers to lead to greater demand for gas-fired electricity.

“It is important that these data centers ensure that the energy they purchase comes from clean renewable energies, such as solar and wind,” says Eikeland. “If we restart fossil fuels to accommodate them, we would be going in the wrong direction.”