Can Innovation Curb AI’s Hunger for Power?
Nvidia says its machine-learning chips have become 45,000 times more energy efficient, with further improvements in the pipeline
6 min
Artificial intelligence is known for its seemingly insatiable appetite for energy. But some tech leaders and analysts are questioning the extent of AI’s footprint going forward, saying innovations in the sector could help offset rising energy demand.
There is certainly no shortage of forecasts detailing the ominous rise in AI’s energy consumption. One commonly-cited report by Climate Action against Disinformation, an association of environmental groups, suggests that AI could drive up global emissions by 80%.
Another estimate by researcher Alex de Vries—which he described as a worst-case scenario—suggests that Google’s AI alone could eventually rack up as much annual electricity demand as the country of Ireland.
Such predictions present a substantial challenge for the relationship between computing and the climate in the coming years.
However, others see reason for optimism. In a Salesforce survey of about 500 corporate sustainability professionals, published Wednesday, nearly half were concerned about AI’s potential negative impacts on sustainability efforts. Meanwhile, almost 60% thought the benefits of AI would offset its risks in addressing the climate crisis.
Microsoft founder Bill Gates recently weighed in on the subject, urging governments not to go “overboard” on concerns about AI’s energy footprint, and suggesting that the technology could actually drive a reduction in global energy demands.
Putting the AI boom in context
The rise of AI should be considered in a broader perspective, according to Charles Boakye, U.S. sustainability analyst at Jefferies. He noted that relative to other industries, the technology still makes up only a fraction of global power demands.
“Data centers—the engines powering everything from AI to traditional computing to cryptocurrency—currently account for about 2% of global electricity consumption. Of this, AI accounts for roughly 0.5%,” Boakye said.
In terms of emissions, statistics last year from the International Energy Agency showed in total, data centers and transmission networks are responsible for 1% of energy-related greenhouse gases. For comparison, the oil-and-gas industry contributed just under 15% of emissions in 2022.
Looking ahead, the intergovernmental organisation said that by 2026, demand for AI is expected to increase ten fold compared with 2023. Even so, IEA data showed that it will still only account for roughly an eighth of total data-center electricity consumption.
“So in terms of AI demand, we’re talking about a small piece of a small piece,” Boakye said, noting that other areas of electrification, such as electric mobility, traditional data center growth and the industrial transition, will ask much greater questions of the power sector.
Charting efficiency trends
Demand is difficult to predict. But recent trends show that in practice, a rise in demand for computing power rarely correlates one for one with a rise in energy consumption, according to climate researcher Jonathan Koomey.
“Between 2010 and 2018, global data centres saw a 550% increase in compute instances and a 2,500% increase in storage capacity. This compares with just a 6% rise in electricity use,” he said.
Koomey, previously a visiting professor at Stanford, Yale and Berkeley, is best known for his work studying long-term trends in the energy-efficiency—now known as Koomey’s Law—highlighting the propensity of computing technologies to become more efficient over time.
AI may be a different animal, with some estimates suggesting large models such as ChatGPT use 10 times more energy than a Google search. But this is unsurprising for a relatively new technology, Koomey noted. In most areas of computing, energy demand tends to spike before levelling off as efficiencies gather pace, he said, noting that forecasts based on this inflection point will often be misguided.
Koomey cited a brief moment in the mid 1990s when internet data flows doubled every hundred days—a statistic that led to overinvestment in networks in the following years and 97% of fibre capacity sitting unused.
Similar efficiency trends can be seen in the development of AI, Jefferies analyst Boakye said. Google’s new TPU processors, for example, are more than 67% more efficient than in 2022, and the energy used to train OpenAI’s ChatGPT models has gone down by 350 times since 2016, he noted.
“It’s in their best interest, and in the interest of their business models, to increase that efficiency,” the analyst said.
Going for gold in the AI Olympics
If any company will have a say in the future of AI, it’s Nvidia . The U.S. technology company designs roughly 80% of the world’s specialized AI chips. Its next-generation GPUs, known as Blackwell, are touted to be 25 times more energy efficient than current iteration Hopper, while offering 30 times more computing power. Blackwell chips are slated for release later this year.
So far, the progress appears consistent, with Hopper 25 to 30 times more efficient than Nvidia’s previous generation of chips. Overall, the company said it has experienced a 45,000 times improvement in GPU energy efficiency over the last eight years.
Nvidia added that software optimisation also plays a big role in increasing the energy efficiency of its products.
“Once a platform has been launched, we will make it more efficient in a single year,” the company said. In the year following its launch in 2022, Hopper became two times more efficient after taking part in MLPerf, otherwise known as the Olympics of AI, in which tech companies compete and collaborate to drive improvements in the speed and efficiency of their models.
Part of this optimization involves taking large, energy-intensive AI models—such as ChatGPT—and refining them to perform more specific tasks. On July 18, for example, OpenAI launched a smaller, smarter and more energy efficient version of its previous GPT model, known as GPT-4o mini. Koomey sees these more “lightweight” models driving demand in the future .
“I’m not convinced large-scale AI has a good business model at this point, despite them driving all the investment. The slam dunk machine learning usually involves smaller, more efficient models, focused on a specific task. For me, this is where most of the business value lies,” he said.
Training and education will be essential for getting a more targeted and efficient experience with AI, helping to narrow the gap between businesses and their sustainability goals, said Suzanne DiBianca, Salesforce’s chief impact officer.
According to Nvidia, another way of alleviating AI’s impact is directing workloads—particularly the more energy-intensive jobs of training AI models—to regions with more abundant resources, ideally renewables which are set to keep a lid on electricity emissions in the coming years.
One company making strides in this space is NexGen Cloud, which builds renewable-powered data centers in areas with untapped energy resources.
According to co-founder and CSO Youlian Tzanev, a large portion of AI workloads don’t need to be performed close to traditional logistical hubs like London or New York, and can be powered from more remote areas in countries like Canada and Norway with excess hydropower.
“We have significantly more power than people believe. The power just isn’t reaching the grid in many cases and is going to waste, and so that is where we focus our efforts,” Tzanev said.
In the U.S., Crusoe Energy Systems offers another example of startups finding innovative ways to power the AI boom. Crusoe’s modular data centers are designed to run on excess natural gas produced at oil wells, achieving a 99.9% methane reduction in the process.
Common forecasting pitfalls
Projecting an outlook for AI’s energy demands is far from straightforward. In its midyear electricity update, the IEA noted that estimates exhibit a wide range of uncertainty, with some analyses following overly “simplistic” extrapolations.
For instance, the organization said certain studies make the mistake of assuming data center operators build all the facilities for which they apply to utilities. Given that several applications can be made for each new data center, this can lead to a multiplication of estimates.
Within data centers themselves, it is tempting for forecasters to imagine computers working flat out around the clock, Koomey said. In practice, GPUs usually operate on much less than their full power capacity, he added.
Based on modeling carried out by Nvidia, GPUs on average tend to run on less than 70% of their potential power. One particular function, known as Multi-Instance GPU, enables workloads—and therefore energy consumption—to be split into seven distinct components, with each able to function independently.
In addition, the company noted the role of substitution effects, in which traditional computing workloads are transferred onto AI platforms and subsequently performed more efficiently—an aspect that can be easily overlooked.
Forecasts can also conflate local data-centre developments with broader energy demands , Koomey added, noting that the most common estimates for electricity demand come from local utility companies.
In the U.S. as a whole, electricity use actually fell in 2023 compared with the previous year, according to the U.S. Energy Information Administration. In March 2024—the most recent month of available data—total demand reached 306 billion kilowatt-hours, down from 317 billion kWh in the year-prior period.
“I worry that people are jumping on the explosive demand-growth train before really understanding what’s going on,” Koomey said. “If you cluster data centers in certain places you’re going to see some local power constraints, but that doesn’t necessarily mean that AI will be a key driver of electricity use more broadly.”
Corrections & Amplifications undefined Nvidia said it has experienced a 45,000 times improvement in GPU energy efficiency over the last eight years. An earlier version of this article incorrectly said the company developed its first GPU eight years ago. (Corrected on July 24)
Copyright 2020, Dow Jones & Company, Inc. All Rights Reserved Worldwide. LEARN MORE
The 1860s Darlinghurst mansion Stoneleigh could become Sydney’s most expensive home ever sold under the hammer when it goes to auction. Clint Ballard is giving buyers a $28 million guide for the heritage-listed mansion on Darley Street, opposite Iona, the former home of Hollywood royalty Baz Luhrmann. Stoneleigh is being offered for sale for the …
Set on one of the city’s last absolute riverfront sites, The Riversdale by Mosaic combines irreplaceable waterfront ownership with one of Brisbane’s most significant residential opportunities.
Gold miners are emerging as a compelling way to navigate market uncertainty, with analysts pointing to strong cash flows, attractive valuations and rising profit margins. As gold prices stabilize above US$4,000 an ounce, mining stocks could offer investors both downside protection and long-term upside.
3 min
Gold is one of the market’s go-to hedges in rocky times. Don’t forget that gold miners’ stocks are too.
The stock market’s gains in 2026 belie the rocky macroeconomic picture: elevated inflation, heightened geopolitical tensions, and jitters about the artificial-intelligence trade. That backdrop, in theory, should be the time for gold to shine. Instead, the price of the yellow metal has tumbled more than 5% so far, after last year’s blistering 65% rally. In part, the U.S. dollar’s recovery has stymied gold, which benefited from the greenback’s weakness in 2025.
Even with the precious metal’s recent weakness, gold mining stocks could be the best way to profit from this year’s uncertainty.
Gold miners “are a valuable hedge against macro risks that would likely be damaging for equities,” BCA Research’s Noah Weisberger and Rishabh Shah wrote this week.
Concerns about the Federal Reserve’s next moves to tackle inflation, the increasingly crowded AI trade, and steep valuations for tech stocks are just some of the drivers that could help gold’s price get on even footing— and lead to even bigger gains for miner stocks.
These stocks’ prices tend to outpace gold’s moves, because the companies have fixed operational costs. So when gold’s price rallies, their profit margins soar, and vice versa. For instance, the VanEck Gold Miners GDX +7.39% exchange-traded fund has fallen 11% this year as the metal has slumped.
Now, gold’s price just needs to stabilize to help miners’ stocks take off, and that seems to be happening. The precious metal has recently found support above the $4,000 level, and has stuck in a narrow range since the end of June. But its price rose ever so slightly in July, ending a four-month losing streak for the metal. Technical analysis also suggests that gold is due for a comeback.
Barron’s recently wrote that the pullbacks for both gold miners and the metal itself are overdone. Senior technical analyst Doug Busch noted that the VanEck ETF is on the “verge of a breakout” and has the potential to hit $11o in early 2027, up more than 40% from its current price.
Gold miners also have more than their role as a market hedge going for them. Their fundamentals are solid, too, says Chris Mancini, portfolio co-manager of the Gabelli Gold Fund.
“Precious metals miners are generating substantial amounts of free cash flow given profit margins of over $2,000 per ounce, and are returning this cash to shareholders through buybacks and dividends,” he said in an email.
“Buying the miners is a cheap way to get exposure to the price of gold,” he added. His fund owns Newmont NEM +6.71%, a Barron’s stock pick last year, and Agnico Eagle Mines as top holdings, as well as miners Northern Star Resources, Endeavour Mining, and Kinross Gold K+8.59%.
Miners are better businesses than they used to be, the BCA team added.
“Capex is more disciplined, margins are high and rising…and they are largely independent of the AI story,” Weisberger, BCA’s head of equities, and Shah, a senior analyst, wrote.
That last part is key. AI is disrupting the software industry and many other services and information-oriented businesses, and investors have piled into AI stocks. But ChatGPT, Claude, Grok, and other large-language models aren’t going to replace the need to mine for metals.
“Equity portfolios can benefit from exposure to quality that is uncorrelated to AI risk, and gold miners fit the bill,” the BCA team said.
They recommend that investors buy the VanEck Gold Miners ETF, which owns top miners such as Agnico, Barrick Mining ABX +7.24%, and Newmont.
An important bonus for big gold miners’ stocks is that their valuations are attractive after the gold’s pullback, too. The VanEck ETF is now trading at just a little more than nine times next year’s earnings estimates. That’s a big discount to its five-year average price-to-earnings ratio of 14, according to FactSet.
What’s more, the ETF is currently valued at a more than 50% discount to the S&P 500 SPX -0.17%, which is trading for about 19 times earnings estimates for 2027. Mining stocks have typically traded at just a 25% discount to the broader market over the past five years. So there is significant upside for the group if valuations move back toward normal levels.
One factor that complicates mining stocks as a market hedge, of course, is if stocks bounce back, which has been the case so far in August.
But both the market and economic outlooks remain cloudy, and investors remain nervous about the Fed’s next moves and AI stocks. Gold miners should do just fine, even if the anxious mood on Wall Street persists.
From Italy’s $93,000-a-night villas to a $20,000 Bowral château, a new global ranking showcases the priciest Airbnbs available in 2026.
Scotch whisky expert, luxury hospitality strategist and Keeper of the Quaich inductee Ross Blainey is bringing a new philosophy of luxury experiences to Citizen Kanebridge.
