Tech vs Carbon: How IBM Achieves Sustainability Goals

As AI adoption and evolution accelerates across the world, concerns of its environmental impact have reared their head.
Whilst companies are becoming more accustomed to the challenge of trying to balance competitive edge against climate commitments, AI’s rapid evolution has brought with it its own sustainability issues.
A primary concern is the growth in large language models (LLMs) and machine learning (ML) systems that have led to increased energy consumption in data centres worldwide – prompting technology leaders to seek solutions that maintain performance while reducing carbon emissions.

This challenge comes as extreme weather events in 2023 caused over US$150bn in damage across the US alone, reminding the world of the urgency for climate action.
Yet unlike other sectors, technology companies now face pressure to develop solutions that help multiple sectors address climate challenges as well as reduce their environmental impacts.

Corporate leaders like Christina Shim, who has been IBM’s Chief Sustainability Officer since May 2024, have inherited these challenges.

IBM deploys AI models to tackle global environmental challenges 

IBM has over 310,000 employees across six continents, which creates significant scope to make a change in climate action.
Yet the company has found that whilst AI is the source of these challenges, it could also be part of the solution.
For example, it is expanding its environmental initiatives through AI and data centre improvements with the company’s Granite AI model, an ML system designed for environmental applications.
One such initiative includes Kenya's national reforestation programme, which aims to plant 15 billion trees in areas where forests support over one-third of national GDP through water retention.

The technology is effective by enabling agricultural companies to predict crop yields during unpredictable weather conditions and helps insurers assess potential fire and flood risks through advanced data analysis.

"This is an incredibly exciting time to lead IBM's strategy around sustainability,” Christina says.
“There is a lot of work ahead in the sustainability field, but I am thrilled to be working on this important issue here.”

Meanwhile, in Denmark, IBM's predictive maintenance technology has extended the lifespan of the Great Belt bridge and tunnel project by 100 years, preventing 750,000 tonnes of carbon dioxide emissions through avoided reconstruction.

Partnerships drive innovation in sustainable technology 

The company collaborates with public, private and non-profit sectors to develop environmental solutions, including serving as "client zero" for new technologies, testing and refining solutions before market deployment.

"At a time when the sustainability landscape is evolving so dynamically, IBM is the perfect place from which to pursue partnerships, drive forward innovation and deploy technology to help accelerate progress," Christina says.

The company operates across 150 countries and has “a century of experience in research, innovation and working with our communities,” she says, leveraging its research capabilities and community engagement to address environmental challenges, including material science innovation and quantum computing development.

Christina notes: “IBM is already playing a part through our AI and energy efficiency work, materials science innovation and thought leadership in quantum commercial development, as a few examples.
“There is plenty of opportunity to do even more, as well as more integrally embedding sustainability throughout the enterprise and how we do business and build products.”

Energy efficiency drives data centre improvements 

IBM sources 74% of electricity for its data centres from renewable sources, with 28 facilities operating on 100% renewable power.

Additionally, its sustainability software has reduced excess computing capacity in AI workloads, cutting energy usage without impacting performance.
Managing to avoid 256,000 megawatt hours of energy consumption, IBM has also implemented over 2,100 conservation projects.
Back in 2023, it further achieved its target of reducing operational greenhouse gas emissions by 65% compared to 2010 levels, two years ahead of schedule.

"Our legacy includes leading the charge in creating environmental policies well before most companies, helping to develop international standards and driving toward ambitious environmental goals in line with the UN IPCC," Christina says.

How AI optimisation reduces environmental impact 

A big contribution to AI’s carbon footprint is the chips behind it, so IBM's researchers have developed AI chip designs that consume 14 times less energy than previous versions.

In August 2024, the company announced new processors for running AI applications on mainframe computers and this optimisation of AI systems extends to the company's foundation models.
For instance, IBM offers Granite in various sizes and employs optimisation techniques to minimise energy consumption during development and deployment.

“Our purpose-built AI models like Granite are helping leaders understand and respond to environmental changes”, Christina says.
“This can help agricultural companies predict crop yields amidst increasingly unpredictable weather, or insurers get a better handle on potential fire or flood risks.

An example is the Government of Kenya using IBM’s geospatial foundation model to help the country fight deforestation and plant 15 billion trees – a critical effort in a country where forested areas retain water and support over a third of the national GDP.”

The initiatives come as extreme weather in the US disrupted supply chains and insurance markets – and Christina notes that this reflects growing awareness of operational risks: “These same challenges reflect an opportunity, as business, government and society are understanding more than before that our ways of operating face a huge, collective risk that must be addressed."

Tech infrastructure improvements target emissions reduction 

Technical infrastructure is one of the most crucial elements to achieving corporate AI sustainability initiatives. 

“There are no shortage of sustainability challenges these days,” Christina says.

“Yet these same challenges reflect an opportunity, as business, government and society are understanding more than before that our ways of operating face a huge, collective risk that must be addressed.”

Through all of IBM’s efforts in AI sustainability, it considers AI ethics as part of every use case that the company’s technology addresses.

IBM believes it can collectively address these problems with AI, rightsizing the models and infrastructure.

“As a leader in AI for business, IBM recognises the critical importance of developing and using these technologies in a sustainable and ethical way,” Christina concludes.

To read the full article in the magazine, click HERE.

Explore the latest edition of Technology Magazine and be part of the conversation at our global conference series, Tech & AI LIVE.

Discover all our upcoming events and secure your tickets today.

Technology Magazine is a BizClik brand