Latest news with #DCByte
Bloomberg
08-05-2025
- Business
- Bloomberg
AI Is Draining Water From Areas That Need It Most
By Michelle Ma Dina Bass Each time you ask an AI chatbot to summarize a lengthy legal document or conjure up a cartoon squirrel wearing glasses, it sends a request to a data center and strains an increasingly scarce resource: water. The data centers that power artificial intelligence consume immense amounts of water to cool hot servers and, indirectly, from the electricity needed to run these facilities. Even before ChatGPT launched in late 2022, communities complained about data centers guzzling up millions of gallons of water every day from cities that didn't have all that much to spare. The problem has only deepened in the years since ChatGPT kicked off an AI frenzy. More than 160 new AI data centers have sprung up across the US in the past three years in places with high competition for scarce water resources, according to a Bloomberg News analysis of data from World Resources Institute, a nonprofit research organization, and market intelligence firm DC Byte. That's a 70% increase from the prior three-year period. As Data Centers Become More Ubiquitous, Water-Stressed Areas See the Most Growth Number of facilities built in the US by year and water stress zone 4008001,2001,600 data centersRelease of ChatGPT A similar trend is unfolding in other countries. Arid regions like Saudi Arabia and the United Arab Emirates are welcoming more data centers than ever before to claim a piece of the AI boom. Meanwhile, in China and India, an even greater proportion of data centers are located in drier areas compared to the US, according to Bloomberg's analysis. That reflects both their growing data center footprint and heightened water scarcity issues caused by climate change. As tech firms have pushed to develop new data centers to support cutting-edge AI systems, they've increasingly turned to states and countries with ample energy resources and favorable regulations, according to experts. What those places often lack, however, is an abundant supply of water. The result is that data centers threaten local water supplies, agriculture and energy production. 'It is very much a growing issue — and it is spreading everywhere,' said Newsha Ajami, chief strategy and development officer for research at Lawrence Berkeley National Laboratory and the founding director of the urban water policy program at Stanford University. In recent years, there have been water-related protests against data centers in the Netherlands, Uruguay and Chile, the last of which temporarily revoked Google's authorization for a planned $200 million facility after people took to the streets. In the US, tech firms have opened more large data centers in drought-prone states like Arizona and Texas, adding to concerns about the local water supply. Data Centers Proliferating Globally in Water-Stressed Areas Number of facilities built in countries with the most data centers in high water risk zones High to extremely high Medium Low China'defined'defined1002003000 Germany0 France0 UK0 India0 Canada0 Australia0 Italy0 Spain0 Netherlands0 Russia0 Indonesia0 SaudiArabia0 Thailand0 UAE0 Chile0 Mexico0 Belgium0 Turkey0 SouthAfrica0 'Every part of the state is facing this water-energy nexus crisis,' said Amy Bush, a hydrologist for RMBJ Geo Inc. based in Abilene, the small Texas city where OpenAI is setting up a massive 1.2-gigawatt data center campus to anchor its $100 billion Stargate AI infrastructure venture. Water is often the last consideration when making siting decisions for data centers because it's cheap compared to the cost of real estate and power, said Sharlene Leurig, a managing member of Fluid Advisors, a water consulting firm. 'Water has traditionally been far less consequential for industrial users than energy costs and availability,' she said. At least one large tech company, Microsoft Corp., is rethinking the pace of data center expansion and pulling back on certain projects. But the number of data centers in dry areas is still poised to increase significantly in the coming years. In the US alone, there are at least 59 additional facilities planned in such regions by 2028, according to DC Byte, with more new projects seemingly getting announced every couple weeks. A Growing Problem More AI means more water. In the US, an average 100-megawatt data center, which uses more power than 75,000 homes combined, also consumes about 2 million liters of water per day, according to an April report on energy and AI from the International Energy Agency (IEA). That's equivalent to the water consumption of about 6,500 households, the report said. Globally, the report estimates data centers consume about 560 billion liters of water annually and that could rise to about 1,200 billion liters by 2030, as tech firms push for bigger facilities stocked with more advanced AI computing chips that run hot. Many data centers rely on evaporative cooling, or 'swamp cooling,' where warm air is drawn through wet pads. Data centers typically evaporate about 80% of the water they draw, discharging 20% back to a wastewater treatment facility, according to Shaolei Ren, an associate professor of electrical and computer engineering at the University of California, Riverside. Residential water usage, by comparison, loses just 10% to evaporation, discharging the other 90%, Ren said. (A spokesperson for Google said the company doesn't have a standard percentage because any data center would see some variation based on factors like location, temperature and humidity.) Read More of Bloomberg's coverage on data centers: AI Is Already Wreaking Havoc on Global Power Systems AI Needs So Much Power, It's Making Yours Worse The water footprint of data centers extends beyond cooling. A 2021 paper found nearly half of US data center servers are fully or partially powered by water-hungry power plants located within water-stressed regions. The IEA estimates 60% of data center water consumption is from indirect use. The surging water demand now threatens two key environmental priorities for tech firms. Before the AI boom, the biggest US cloud computing companies — Inc., Microsoft and Google — pledged to cut their carbon footprint in the coming years and also be water positive by 2030, meaning they would add more water to the environment than they use. However, the regions with the most available renewable energy resources to support data centers — particularly solar — are sometimes the ones with the least water. Conversely, data centers set up to use less water in hotter regions end up requiring more power to run. 'When water has this power of reducing energy consumption and energy-related emissions like carbon, then we have to be able to discuss the role that it could, should, or should not play in different applications,' Ben Townsend, global head of infrastructure and sustainability at Google, said in an interview. What Water-Stressed Sites for Data Centers Look Like Satellite imagery showing the surrounding areas of planned and existing data centers in a selection of places where water stress is high or extremely high Searching for a Solution Tech firms have faced these opposing priorities before. Early in the history of data centers, companies relied on air conditioning to cool equipment. These systems used a lot of electricity, causing companies to look to water to create more energy-efficient cooling technology, said Steve Solomon, vice president of datacenter infrastructure engineering at Microsoft. Now tech companies are trying new solutions, including data center and chip designs that let them use less water. Some are placing hot chips directly on cold plates that use water or else submerging chips and servers in liquid, a process known as immersion cooling. Businesses are also experimenting with synthetic liquids to cool data centers. But some coolants are being phased out from the market because they use so-called forever chemicals, which don't naturally break down and can persist in animals, people and the environment. Recently, Microsoft said it developed a data center design that is closed so that water doesn't evaporate but rather is constantly circulated between servers and chillers, without the need for refilling. The design will be deployed first in facilities in Wisconsin and Arizona, planned for 2026. Crusoe Energy Systems, a developer behind OpenAI's Stargate site in Abilene, also plans to use closed-loop cooling systems. But here, too, 'there is a tradeoff in energy,' said Ben Kortlang, a partner at G2 Venture Partners, an investor in Crusoe. These systems are more power-hungry than evaporative methods, he said. For all the research into alternative approaches, a majority of AI-specialized data centers used evaporation-based cooling systems either 24/7 or during hot days in 2023, with more data centers expected to use water evaporation across the entire industry by 2028, according to a Lawrence Berkeley National Laboratory report released late last year. Amazon Web Services, for example, relies widely on direct evaporative cooling, the company said. AWS does turn to different cooling methods based on the time of the year and uses sensors to react to changing weather. To minimize impacts on drinking water, Amazon's cloud unit cools 24 of its data centers using treated sewage. The company said it is hoping to dramatically expand that in coming years by working with local utilities to access the recycled water it needs. The cloud computing firms also work with local nonprofits and officials on watershed restoration, replenishment and other projects to return water. Lack of Transparency As Silicon Valley mulls solutions, water advocates say tech companies need to be more transparent about the problem. Almost no information about data center water usage on an individual system level is publicly available. In one case, The Dalles, a city in Oregon, sued Oregonian Media Group to prevent the release of Google's water usage records, arguing the information was a trade secret. The city agreed to disclose those records after a 13-month legal battle. (Google's Townsend said the company contributed to the available water in The Dalles by developing water resources at the site that were previously unavailable and giving some of that back to the community to 'increase the value available to everyone.') Jennifer Walker, director of the Texas Coast and Water Program at the National Wildlife Federation, also said state officials need more information for water planning. But when the Texas Water Development Board sent a water use survey to data centers, it received a lackluster response, she said. 'We just had one of the hottest summers on record in Texas, and we've had several of those,' she said. 'I'm concerned about any super water-intensive industry that is going to come into our state.' Related tickers: 1554630D:US (OpenAI Inc.) AMZN:US ( Inc.) MSFT:US (Microsoft) GOOGL:US (Alphabet Inc.) Edited by Seth Fiegerman Chloe Whiteaker Brian Eckhouse With assistance from Rachel Metz Brody Ford Jeremy Diamond Leon Yin Methodology Bloomberg analyzed "Baseline Water Stress" data from the World Resources Institute's Water Risk Atlas. Baseline water stress is recorded in five categories: "low", "low - medium", "medium - high", "high", "extremely high", "arid and low water use". For the purposes of this analysis, Bloomberg combined "low-medium" and "medium-high" into one category ("medium"), and combined "high", "extremely high" and "arid and low water use" into one category ("high to extremely high"). Bloomberg analyzed data center locations and capacity — including active facilities, planned facilities, and those under construction — compiled by DCByte, a market intelligence firm. In order to account for data center that may be located at the edge of multiple water stress areas with different risk levels, Bloomberg used a CDKTree algorithm to find all of the water stress areas within a 30 mile radius of each data center location. For each data center, Bloomberg assigned the maximum water stress score found within this search radius. Bloomberg calculated the total number of data centers that fell within each water stress category.
The Sun
05-05-2025
- Business
- The Sun
Meeting the AI challenge: Six practical, green approaches
'THE best time to plant a tree was 20 years ago, and the second-best time is today.' This saying is more relevant than ever in the race to develop sustainable data centres, especially with the rapid rise of AI. This transformative technology is driving energy demand to unprecedented levels, threatening global sustainability goals. AI mirroring energy use The numbers tell a sobering story. The rise of AI is fueling demand for high-volume processing, with data centres projected to consume 2% of global electricity in 2025 — about 536 TWh — doubling to 1,065 TWh by 2030. Malaysia is fast becoming a heavyweight in the data centre arena, and Johor exemplifies this rapid expansion. In early 2021, the state's data centre capacity stood at just 10MW. By 2024, this capacity has surged to over 1,500MW, according to DC Byte. Generative AI is the driving force behind this expansion, with investment in mission-critical infrastructure expected to rise from US$125.35 billion (RM547 billion) in 2024 to US$364.62 billion by 2034. While this growth unlocks new possibilities across the world, but also raises pressing sustainability concerns, requiring a fundamental shift in how data centres source, manage, and consume energy. What's at stake AI is reshaping industries and is projected to add US$15.7 trillion to the global economy by 2030. However, it's a double-edged sword. Malaysia's Natural Resources and Environmental Sustainability Minister Nik Nazmi Nik Ahmad has cautioned against these environmental impacts, prompting the government to adopt a more selective approach to tech investments. Johor's AI-driven data centres illustrate this concern, with projections of electricity demand hitting over 5,000 MW by 2035 and applications for supply already surpassing 11,000 MW — nearly 40% of Peninsular Malaysia's power capacity. While AI increases productivity, it also intensifies power consumption, heat generation, and emissions, requiring advanced cooling solutions. Simultaneously, data centre infrastructure management solutions integrated with AI analytical power can proactively simulate and address enterprise data centre challenges. In the face of growing global attention on sustainability, inaction is no longer an option. With Malaysia's focus on sustainability, the industry has a clear mandate: embrace sustainability and efficiency as environmental imperatives as well as smart business strategies. The good news is there are tremendous benefits to this approach. Six practical solutions Above all else, a commitment to sustainability must be practical. To curb AI's environmental impact, operators must tackle three core challenges: strengthening sustainability strategies, addressing Scope 3 emissions, and meeting regulatory demands. Here's how: 0 Evaluate infrastructure for efficiency: Identifying inefficiencies in cooling systems, server utilisation, and energy management for significant improvements. 0 Adopt efficient infrastructure: Transitioning to modern servers, advanced cooling systems and renewable energy sources to reduce carbon footprint. Designing facilities with energy efficiency in mind and retrofitting older infrastructures can further optimize energy usage and contribute to substantial savings. 0 Collaborate with sustainable suppliers: Sustainability isn't achievable in isolation. By collaborating with suppliers committed to decarbonising their supply chains, operators can effectively address Scope 3 emissions, the largest source of greenhouse gas emissions for businesses. 0 Leverage advanced reporting tools: Real-time monitoring enables operators to identify inefficiencies and implement corrective measures promptly, optimising energy use. 0 Leverage AI for optimisation: AI can play a significant role in efficiently allocating resources, predicting maintenance requirements, and integrating renewable energy — ultimately minimising waste and lowering emissions. 0 Continuously reassess and improve: Sustainability is an ongoing effort — regular assessments, benchmarking against industry standards, and adopting best practices will drive lasting progress. Second best time You can't go back in time to plant a tree, but you can do the next best thing and plant one today. The time to rethink your sustainability agenda is now. As AI accelerates growth in the data centre industry, it is crucial to take immediate action to rein in carbon emissions. While expertise and resources may be limited, companies can collaborate with vendors offering sustainability services to craft comprehensive strategies for a greener future. This article is contributed by Schneider Electric country president for Malaysia, Eugene Quah.
Yahoo
18-04-2025
- Business
- Yahoo
Malaysia aspires to provide the backbone of the global AI boom. Will the bet pay off?
The jungles of Johor, the Malaysian state across the Johor Strait from Singapore, were first cleared in the 1840s by Chinese clans from Singapore seeking more space to grow black pepper. In the next century, under British rule, those pepper farms gave way to vast plantations of rubber and oil palm trees. On many of those same sites today, Johor is cultivating a new kind of cash crop: data centers meant to feed the world's voracious appetite for artificial intelligence. Johor's data center boom, like the shift to growing pepper, is partly a function of scarcity in Singapore. The tiny city-state is Southeast Asia's digital hub. But it imports both water and power, and in 2019 imposed a moratorium on building data centers because the hulking facilities were guzzling water and consuming 7% of Singapore's electricity. Investors and operators of data centers flocked to neighboring Malaysia, where land is cheap, energy is abundant, and the government is eager to jump-start development of the nation's digital economy. But Johor's rise as a data center powerhouse is also driven by the global scramble for computational power. Singapore rescinded its data center ban in January 2022, but the release of ChatGPT later that year triggered an explosion in global demand for AI infrastructure—and ignited a new investment frenzy in Malaysia. In 2023, Malaysia reaped more than $10 billion in investments for data centers, then tripled that in 2024, making the country the world's hottest destination for data center investments in both years, according to property consultancy Knight Frank. Johor is the epicenter of that construction surge. For the state, and for Malaysia, the big question is whether this flood of capital and expertise will carry their broader economies to a new era of high-tech growth—or whether other challenges, like shifts in global demand and constraints in local resources, will turn their data centers from cash cows to liabilities. Johor hosts more than 40 data centers that are either operational or under construction, according to advisory firm Baxtel, up from about a dozen in 2022. Many more are in the planning stages. Data center capacity, measured in how much power the facilities can provide, surged to over 1,500 megawatts last year, up from 10 megawatts three years earlier, according to data center market intelligence platform DC Byte. If expansion continues at its current breakneck pace, Johor could overtake Northern Virginia as the world's largest data center corridor within the next five years. 'Johor is adding data center capacity at a speed and scale I've not seen ever anywhere else in the world,' says Rangu Salgame, CEO of Princeton Digital Group, a Singapore-based data center operator. Princeton Digital, whose backers include private equity giant Warburg Pincus, last year launched the first phase of a $1.5 billion, 150-megawatt data center campus in a massive tech park 40 miles inland from the Singapore-Johor border crossing—and plans to add a second, 200-megawatt campus at a business park a few miles up the road. The parade of companies piling in with multibillion-dollar investment announcements in Johor also includes global tech leaders like Nvidia, Microsoft, Alphabet, and Oracle, plus data center operators such as California's Equinix, Japan's NTT Data, and China's GDS Holdings. 'Three years ago,' says Salgame, 'if you'd asked CEOs of the global tech giants about Johor, they'd have never heard of it, much less be able to find it on a map. Now, everyone is here.' It's no accident that the advent of large language models (LLMs) in the U.S. and China has sparked a data center bonanza in faraway Malaysia. In the pre-ChatGPT era, for many of the services handled by data centers, there was an enormous advantage to operating from facilities physically close to end users. For functions like online gaming, stock trading, fraud detection, social media, or streaming videos, every millisecond counts. Companies providing such services pay huge penalties for 'latency'—sluggishness in the time it takes for data to travel between a user's device and the data center and back. In contrast, training LLMs isn't interactive. Instead of sending requests and waiting for real-time responses, it involves running long, continuous computations on fixed datasets. The process can run for days or weeks without needing rapid back-and-forth communication. When latency isn't a concern, AI companies can instead prioritize efficiency—cheap and abundant power and land—and locate data centers thousands of miles from where the models are designed or meant to be used. That means Malaysia's AI data centers can compete not only with those in Singapore or other Southeast Asian neighbors, but also with similar facilities worldwide. Malaysian Prime Minister Anwar Ibrahim has welcomed the data center boom and is rolling out strategic initiatives, including tax breaks and streamlined approval procedures, to position the nation as a global AI hub. A critical part of that push is the Green Lane Pathway, a 2023 initiative launched by Tenaga Nasional Berhad, Malaysia's primary electricity utility, that aims to reduce the time required to connect data centers to the power grid to 12 months, down from more than three years prior. There are signs the data center boom is straining Malaysia's resources—for some of the same reasons the facilities were temporarily banned in Singapore. Malaysia, like Singapore, is one of the most water stressed countries in the world. Malaysia's National Water Services Commission has warned that the country could face widespread water shortages in the next five years owing to climate change and aging infrastructure—even without factoring in increased demand from data centers. Power, too, is an issue. A medium-size data center might have a capacity of 40 to 50 megawatts, enough to consume as much electricity in a year as about 125,000 homes, depending on usage. Large hyperscale AI processing centers can require as much as 500 megawatts continuously, consuming more electricity annually than the roughly 250,000 households in Johor's largest city, Johor Bahru. Malaysia's position on the equator means that its data centers also require far more energy to cool than facilities in northern countries with colder climates. At a recent investor conference, Johor Bahru Mayor Mohd Noorazam Osman acknowledged concerns about water and power shortages. 'People are too hyped up about data centers nowadays,' he said. 'The issue in Johor is we do not have enough water and power. I believe that while promoting investments is important, it should not come at the expense of local and domestic needs of the people.' The Malaysian government says it expects data centers operating in the country to pay a premium for water and power; early indications suggest tech companies and operators are willing to do so. Authorities last year rejected the applications of a handful of data center projects for failure to comply with efficiency and sustainability standards. The rapid increase in power demand from data centers could prove a boon if it accelerates Malaysia's transition to renewable energy. In 2020, only about 4% of Malaysia's power came from renewable sources. That percentage is expected to rise above 30% this year, according to the Malaysian Investment Development Authority, and the government has vowed to increase the share of renewable energy in total generation capacity to 70% by 2050. In Johor, authorities are thinking big. Princeton Digital's first data center campus is located in Sedenak Tech Park (STeP), a 700-acre digital hub owned by the property arm of JCorp, a state-owned conglomerate, on a site that was once part of a sprawling palm oil plantation. In addition to the Princeton hub, STeP includes a 300-megawatt hyperscale data center campus being built by Amsterdam's Yondr Group, and a third, under development by Japan's Mitsui & Co., that will include an on-site solar farm. STeP, already Malaysia's largest data center complex, is about to get bigger. JCorp is developing a second phase, STeP 2, that will add another 640 acres to the park, and has plans for a 7,000-acre township that will include R&D facilities, residential areas, and culture and rec centers. JCorp also has engaged Zaha Hadid Architects to design a 500-acre innovation hub called Discovery City that will integrate digital technologies and sustainable living. The proliferation of such projects is transforming Johor, Malaysia's southernmost state. Johor and Singapore are connected by two land crossings, Woodlands and Tuas Link, that are among the busiest and most congested border crossings in the world. The Singapore side is densely populated and carefully organized, with tolls and customs digitized. The Johor side is bustling and chaotic, with far more motorcycles, small cars, and buses. "Johor is adding data center capacity at a speed and scale I've not seen ever anywhere else." In January, Johor signed a 'special economic zone' agreement with Singapore to promote cross-border cooperation between the two economies. The motorbike-loving billionaire sultan of Johor, currently Malaysia's king under the country's rotating monarchy system, is championing the effort to bring his state and Singapore closer together. The agreement includes tax breaks, enables smoother cross-border trade, and makes it easier for skilled labor to move back and forth across the border. It's unclear whether data centers will generate more and better jobs for Johor. Most data centers provide about 30 to 50 permanent jobs. Larger facilities might create as many as 200. But on their own, data centers seem unlikely to significantly boost overall wealth in Johor, where the GDP per capita is about $10,000 compared with nearly $85,000 in Singapore. Nor is it clear that Malaysia can use the development of data centers to attract other tech industries such as chip manufacturing. The larger risk is a global data center bubble. The so-called DeepSeek Shock (China's breakthrough AI model that rattled Wall Street) could reduce the scale and demand for data centers everywhere if an overhaul of AI models to match DeepSeek's low-cost approach reduces demand for cutting-edge chips, expanded power plants, and large-scale data centers. Salgame, for his part, says he is 'not the least bit worried' about flagging demand for computational power from the data centers Princeton Digital is building in Johor. Cheaper, more efficient AI models will only accelerate the world's use of AI—and the need for low-cost AI training centers in places like Johor. 'This is only the beginning,' he says. This article appears in the April/May 2025: Asia issue of Fortune with the headline "Malaysia's data center power play." This story was originally featured on
Yahoo
28-01-2025
- Business
- Yahoo
DeepSeek shock ‘puts Starmer's data centre plans at risk'
Sir Keir Starmer's plan to build scores of new data centres in Britain to make the country a 'world leader' in artificial intelligence (AI) has been thrown into doubt by the emergence of highly efficient Chinese chatbot. Experts said two fifths of the data centres being planned in Britain were now at risk of delay or cancellation as a result of the debut of DeepSeek, a Chinese AI that has sent shock waves through the technology industry. The highly sophisticated model has been built for a fraction of the cost of US rivals and using far less computing power, raising questions about how many data centres will ultimately be required for AI. Consultancy DC Byte said the DeepSeek scare was likely to shake confidence in data centre proposals that were not yet 'firmed up', or ones that were centred solely around AI. Data centres that would consume around 4,100 megawatts (MW) of power are currently in the pipeline across the country, compared to around 1,500MW online today. The enormous construction boom has been accelerated by expectations that powerful AI software will demand ever greater amounts of computing power to develop. But on Tuesday, DC Byte warned that around 1,700MW, or 41pc, of the extra capacity being planned was considered 'early stage' and now looked less certain. Edward Galvin, the chief executive of DC Byte, said: 'It's those companies that are developing data centre capacity or power generation capacity just for AI that will definitely be suffering now. 'With those early-stage schemes, there is no firm commitment and nothing is under construction yet, so it is much easier for people to back out.' Mr Galvin said he expected rents to fall on some data centres and for some early-stage deals to be put on the shelf for the time being. Doubts over data centre projects follows revelations that DeepSeek has developed a ground-breaking piece of AI software that uses just 10pc of the computing power of comparable models created by US rivals such as OpenAI and Meta. The announcement has shaken Silicon Valley executives and sent shares in tech companies, data centre developers, energy suppliers and nuclear power companies – all of which are expecting to benefit from an AI boom – plunging on Monday. DeepSeek has also cast doubt over Britain's AI action plan. The Prime Minister last month announced the creation of 'AI Growth Zones' to speed up planning approvals for data centres, part of attempts to make sure Britain does not fall behind in the global technology race. Despite doubts over AI-focused projects, Mr Galvin said demand for data centres more generally remained solid as companies and governments shift more data to the 'cloud' and as the use of social media and video streaming services such as Netflix continues to rise. Mr Galvin said: 'Is there fundamentally still demand? Yes. Was it all getting a little bit silly and a little bit frothy? Probably, yes.' Cameron Bell, of estate agent Savills, said: 'In truth, it's too early to say [whether DeepSeek will have an effect]. But by all accounts, we've been told to continue the work that we're doing. We haven't had any deals fall out of bed yet as a result of this.' Sign in to access your portfolio
Yahoo
28-01-2025
- Business
- Yahoo
DeepSeek shock ‘puts Starmer's data centre plans at risk'
Sir Keir Starmer's plan to build scores of new data centres in Britain to make the country a 'world leader' in artificial intelligence (AI) has been thrown into doubt by the emergence of highly efficient Chinese chatbot. Experts said two fifths of the data centres being planned in Britain were now at risk of delay or cancellation as a result of the debut of DeepSeek, a Chinese AI that has sent shock waves through the technology industry. The highly sophisticated model has been built for a fraction of the cost of US rivals and using far less computing power, raising questions about how many data centres will ultimately be required for AI. Consultancy DC Byte said the DeepSeek scare was likely to shake confidence in data centre proposals that were not yet 'firmed up', or ones that were centred solely around AI. Data centres that would consume around 4,100 megawatts (MW) of power are currently in the pipeline across the country, compared to around 1,500MW online today. The enormous construction boom has been accelerated by expectations that powerful AI software will demand ever greater amounts of computing power to develop. But on Tuesday, DC Byte warned that around 1,700MW, or 41pc, of the extra capacity being planned was considered 'early stage' and now looked less certain. Edward Galvin, the chief executive of DC Byte, said: 'It's those companies that are developing data centre capacity or power generation capacity just for AI that will definitely be suffering now. 'With those early-stage schemes, there is no firm commitment and nothing is under construction yet, so it is much easier for people to back out.' Mr Galvin said he expected rents to fall on some data centres and for some early-stage deals to be put on the shelf for the time being. Doubts over data centre projects follows revelations that DeepSeek has developed a ground-breaking piece of AI software that uses just 10pc of the computing power of comparable models created by US rivals such as OpenAI and Meta. The announcement has shaken Silicon Valley executives and sent shares in tech companies, data centre developers, energy suppliers and nuclear power companies – all of which are expecting to benefit from an AI boom – plunging on Monday. DeepSeek has also cast doubt over Britain's AI action plan. The Prime Minister last month announced the creation of 'AI Growth Zones' to speed up planning approvals for data centres, part of attempts to make sure Britain does not fall behind in the global technology race. Despite doubts over AI-focused projects, Mr Galvin said demand for data centres more generally remained solid as companies and governments shift more data to the 'cloud' and as the use of social media and video streaming services such as Netflix continues to rise. Mr Galvin said: 'Is there fundamentally still demand? Yes. Was it all getting a little bit silly and a little bit frothy? Probably, yes.' Cameron Bell, of estate agent Savills, said: 'In truth, it's too early to say [whether DeepSeek will have an effect]. But by all accounts, we've been told to continue the work that we're doing. We haven't had any deals fall out of bed yet as a result of this.' Broaden your horizons with award-winning British journalism. Try The Telegraph free for 1 month with unlimited access to our award-winning website, exclusive app, money-saving offers and more.
