It requires a huge amount of energy to operate the data center minds of widely used AI models. The demand for this energy is only getting bigger. By 2024, many of the biggest technology companies in Silicon Valley and numerous well-funded startups have publicly committed to climate action, emphasizing their sustainability goals and their pledges to be carbon neutral, as well as their commitment to using recycled materialsmaterials. However, as the energy-intensive nature of AI becomes more evident, it seems that many of those environmental priorities could be at risk.
According to a forecast from the International Energy Agency in March, the energy-hungry AI models and cryptocurrency mining combined could double the energy consumption of data centers worldwide in just two years . Recent reports indicate that tech leaders looking to stay competitive in the growing AI industry may be considering turning to traditional, carbon-emitting energy sources to meet this demand.Data centers need more energy to support AI models
Even though the precise measurement of AI's energy consumption is still debated, it is increasingly evident that the complex data centers required to train and operate these systems consume a lot of energy. According to a recently published peer-reviewed data analysis,
the energy demand from AI servers in 2027 could be equivalent to that of Argentina, the Netherlands, or Sweden combined . The construction of new data centers is also not slowing down. According to a report by the Washington Square Journal,Amazon Web Service Vice President of Engineering Bill Vass stated at an energy industry event in Texas that a new data center is being built every three days. Other energy industry leaders, such as former U.S. Energy Secretary Ernest Moniz, expressed concerns that renewable energy production might not be sufficient to power the projected growth of data centers. According to Moniz, 'We're not going to build 100 gigawatts of new renewables in a few years.' He further stated that meeting the unmet energy demands brought on by AI, primarily through electricity, would require tapping into more natural gas and coal power plants. He also mentioned that when it comes to meeting energy demands with renewables, 'you're kind of stuck.' Meanwhile, Dominion Energy CEO Robert Blue mentioned that the increased energy demand has led them to build a new gas power plant while working towards achieving a net-zero goal by 2050. Additionally, executives from natural gas companies stated that tech firms expanding their data centers have shown interest in using natural gas as an energy source.
Tech companies have a mixed track record when it comes to fulfilling sustainability promises
A sudden reconsideration of non-renewable energy sources to power an Journalcould contradict the net-zero carbon timelines and sustainability commitments made by major tech companies in recent years. Microsoft and Google, who are competing in the rapidly evolving generative AI tools like ChatGPT and Gemini, have both outlined plans to achieve net
negative emissions in the near future
. As for Apple, which AI boom reportedly ended its long-standing car unit to focus on AI instead, wants to becomecarbon neutral in its global supply chains by 2030 . The Biden administration has pledged ambitiously the US to have a carbon pollution free electricity sector by 2035.Some companies with ‘net-zero’ goals have been criticized for greenwashing Critics argue that some climate pledges, especially those from large tech firms, may seem impressive on paper but have fallen short in key areas. Multiple independent monitors have criticized large tech companies in recent years for allegedly not properly disclosing their greenhouse gas emissions
[ Related: . Others have criticized tech firms for heavily basing their sustainability ]
strategies on carbon offsets instead of potentially more effective solutions like reducing energy consumption. The enticing race for AI dominance risks stretching those already strained goals even further.The AI boom has led to new data centers popping up across the US Demand for electricity is increasing across the country. According to a report in the Washington Post
, expected energy production in Georgia in the next decade is 17 times larger than it was recently. Northern Virginia, according to the same report, could require the equivalent of several nuclear power plants to meet the increased demand from planned data centers currently under construction. New data centers have appeared in both of those states in recent years. Lobbyists representing traditional coal and gas energy providers are also urging government offices to delay retiring some fossil fuel plants in order to meet increasing energy demands. Data centers in the US alone were responsible for 4% of the country’s overall energy use in 2022
according to the IEA. That figure will only grow as more AI-focused facilities come online. a recent At the same time, some of the strongest proponents of the AI industry have argued that these energy-intensive models may be instrumental in helping to scale up renewable energy sources Post and develop technologies to counteract the most destructive aspects of climate change. Previous reports suggest that powerful AI models could improve the efficiency of oil and gas facilities by enhancing underground mapping. AI simulation models could also help engineers create optimal designs for wind or solar plants that could reduce their cost and make them more desirable as an energy source. Microsoft, which partners with OpenAI, is reportedly already using generative AI tools to try and
simplify the regulatory approval process for nuclear reactors . Those future reactors, in theory, would then be used to generate the electricity needed to power its AI models. Fossil-fueled AI prioritizes long-term optimism over current-day climate realities argue The issue with those more optimistic outlooks is that they remain, for the time being at least, mostly theoretical and severely lacking in real-world data. AI models improve the efficiency and affordability of renewable resources in the long term, but they risk doing so by increasing the use of non-renewable resources right now. And with other industries are experiencing a sudden increase in energy demands at the same time, these positive long-term forecasts could justify spending on natural gas and coal in the short term. Supporting all of this is a worsening climate outlook that most climate scientists and international organizations agree requires drastic action to reduce emissions as soon as possible. Renewable energy sources are gaining popularity in the USin the US, renewable energy sources are increasing
but technology companies seeking more easily accessible electricity for their upcoming AI projects may hinder that progress.
Recent studies indicate that renewable energy sources alone will not be sufficient to meet the increasingly high power needs of data centers. may increase the efficiency and affordability of renewable resources long term, but they risk doing so by pushing down on the accelerator of non-renewable resources right now. And with energy demands surging in other industries outside of tech at the same time, these optimistic longer-term outlooks could serve to justify splurging on natural gas and goal in the short term. Underpinning all of this is a worsening climate outlook that the overwhelming majority of climate scientists and international organizations agree demands radical action to reduce emissions as soon as possible. Renewable energy sources are on the rise in the US but tech firms looking for easier available sources of electricity to power their next AI projects risk setting back that progress.
