In the future, historians will either look back at 23 November 2022 as the day humanity secured its future or sealed its fate. This was the day OpenAI launched ChatGPT. In the two years since, some of the world’s largest businesses have invested billions in artificial intelligence (AI). Leaders have predicted AI will improve efficiency, increase productivity, reduce costs, relieve workers of tedious jobs and power economic growth. But at what cost?
A 2023 study by Cornell University revealed that ChatGPT uses nearly 500ml of water for every 50 prompts. Research carried out by Goldman Sachs also found that ChatGPT needs nearly 10 times as much electricity as the Google search engine to process a query. At present, data centres worldwide consume 1% to 2% of overall global power usage, but this is predicted to rise to 3% to 4% by the end of the decade as AI use increases.
If these predictions come true, carbon emissions from data centres could double by 2030, meaning businesses will surely fail to meet the UK government’s legal target of reaching net zero by 2050. Despite this, the integration of AI in businesses is on an unstoppable rise. In a 2024 survey by Global Data, around 27% of businesses report a high adoption rate of AI in their workload processes. This is up from just 10% in November 2023.
A greener future
So with AI here to stay and businesses’ sustainability targets written into UK law, leaders are faced with an undeniable realisation: the need for change is paramount. But it’s not all bad news for those working towards a tech-driven, sustainable world.
In fact, in the future AI could be used to aid businesses’ sustainability efforts in many ways. Take for instance carbon accounting, the process of measuring and tracking the amount of greenhouse gases produced by an organisation.
This currently requires leaders to grapple with complex calculations to produce accurate and up-to-date figures. But AI could help to create more intelligent and efficient readings. Energy trading, the buying and selling of energy commodities, could also be enhanced with AI, enabling businesses to navigate volatile prices and save money.
But it’s the algorithms themselves, which serve as the brains of data centres, that must be targeted for AI to have a sustainable future. “The training phase of machine-learning models requires huge amounts of energy,” says Sheila Rohra, CEO at Hitachi Vantara, a global leader in building advanced data infrastructure, intelligent data management and AI-powered hybrid-cloud solutions for businesses.
“If businesses routinely retrain these models to adapt to evolving business needs, their energy needs will be substantial and long-term. This simply won’t be sustainable with their ESG goals. We need to retune algorithms so they require less retraining and energy consumption.”
Smarter algorithms, housed in energy-efficient data-storage platforms, can enable leaders to decarbonise their data centres at speed. As a result, organisations can arm themselves with the AI-driven insights they need to decarbonise their wider operations and achieve their sustainability goals.
AI-powered sustainability
But focusing on one solution in isolation won’t be enough to achieve sustainable AI. Instead, Hitachi Vantara uses a wider, AI-focused strategy to identify inefficiencies across businesses’ entire operations. Its AI discovery service generates data-driven insights to inform key business decisions. This includes assisting in the creation of predictive models to optimise processes, reduce costs and improve operational efficiency, particularly within areas such as manufacturing, logistics and energy management.
With AI here to stay and businesses’ sustainability targets written into UK law, the need for change is paramount
Rohra says the AI discovery service is already helping customers to overcome several common sustainability bottlenecks. “Some businesses lack the expertise and specialised knowledge to transition to more energy-efficient operations,” she says. “They may also be reluctant to invest vast sums of money upfront into new energy-efficient technologies. AI can simplify this process by providing intelligent insights into the short- and long-term sustainability and operational benefits and reduce costs.”
Organisations are already benefitting from these AI-driven insights. Infosys managed to reduce its electricity and air-conditioning costs by 60% by using Hitachi Vantara’s technology in its data centres. It achieved this while continuing to run applications up to 90% faster than it could previously. The organisation also delivered a seamless storage migration for BMW Group, which led to a 70% smaller carbon footprint and a data centre that requires 80% less power, all without disrupting business operations.
Decarbonising data centres
Another crucial element of making tech sustainable at a global level lies in the decarbonisation of data centres. Despite the growing energy consumption of data centres, Rohra firmly believes that technology can also be part of the sustainability solution. Hitachi Vantara is on a mission to help businesses achieve sustainable AI and has set its own ambitious target to become carbon neutral in direct and indirect emissions by 2030.
“We understand it has to start with us,” says Rohra. “Transparency is essential to instilling confidence in our ESG commitments and we are proud to say our technology has one of the lowest carbon footprints compared to other data-infrastructure providers.” She says it’s possible for businesses to increase AI adoption sustainably with the correct infrastructure in place. “We are able to deliver systems and solutions that can actually help drive other businesses to be more energy efficient,” she adds.
Hitachi Vantara is helping businesses to decarbonise their data centres through the adoption of energy-efficient data storage technology. Its virtual-storage platform one block has been shown to reduce data-centre power consumption by 30% to 40%. It also earned all three of the top rankings for ‘best storage solutions available’ by Energy Star® certification. Gijima, a South African information and communications tech company, deployed the platform to consolidate its physical infrastructure, which led to a 66% reduction in power consumption and a reduction in CO2 emissions.
The journey towards sustainable AI is not without its challenges, but it’s also full of opportunities. By prioritising smarter algorithms, energy-efficient infrastructure and holistic strategies, businesses can strike a balance between technological advances and environmental responsibility. The future of AI doesn’t have to come at the expense of the planet. If innovation is matched with action, sustainability can thrive alongside progress.
For more information please visit hitachivantara.com
In the future, historians will either look back at 23 November 2022 as the day humanity secured its future or sealed its fate. This was the day OpenAI launched ChatGPT. In the two years since, some of the world’s largest businesses have invested billions in artificial intelligence (AI). Leaders have predicted AI will improve efficiency, increase productivity, reduce costs, relieve workers of tedious jobs and power economic growth. But at what cost?
A 2023 study by Cornell University revealed that ChatGPT uses nearly 500ml of water for every 50 prompts. Research carried out by Goldman Sachs also found that ChatGPT needs nearly 10 times as much electricity as the Google search engine to process a query. At present, data centres worldwide consume 1% to 2% of overall global power usage, but this is predicted to rise to 3% to 4% by the end of the decade as AI use increases.
If these predictions come true, carbon emissions from data centres could double by 2030, meaning businesses will surely fail to meet the UK government’s legal target of reaching net zero by 2050. Despite this, the integration of AI in businesses is on an unstoppable rise. In a 2024 survey by Global Data, around 27% of businesses report a high adoption rate of AI in their workload processes. This is up from just 10% in November 2023.
