Greening AI

Greening AI

Welcome to Institute Insights, where TBI experts bring to life our work supporting political leaders to drive change that transforms lives.?

In this edition, Senior Policy Advisor Devorah West explores how governments have an opportunity to drive twin transitions in AI development and clean-energy advancements while Global Government Engagement Advisor Marie Teo looks at the lessons that can be learned from Singapores global leadership in AI and digital sustainability.

If you want to know more about our work, subscribe to the TBI newsletter .

Unlocking the Twin Transitions in AI and Energy

Over the past several months, it has been hard to escape headlines pointing to the vast amounts of energy artificial intelligence consumes and the stress this is putting on national grids. While some reports are decidedly overblown, there is no doubt that AI has an energy problem.

Training the AI model GPT-3 is estimated to have emitted as much as 552 tonnes of carbon-dioxide equivalent (CO2e) – this is around the same level of emissions as 200 UK homes for one year.?Data centres (excluding those for cryptocurrency) currently account for 1 to 1.5 per cent of global electricity use. The International Energy Agency (IEA) projects that global electricity demand from data centres could double between 2022 and 2026.

Around the world, this is undermining attempts to hit net-zero targets as countries meet spikes in demand by turning to fossil fuels. This increased energy consumption cannot be entirely offset by the energy-efficient improvements AI will make. Instead, governments will need to take action across AI’s value chain, from the grid and power generation to AI’s hardware, software and end use.

But rather than see this as a challenge, governments have an opportunity to drive AI development and clean-energy advancements together . In fact, as AI becomes more integral to the modern economy, countries will not be able to achieve one without the other. The AI revolution and the energy transition have become deeply intertwined, with each driving the success of the other.

The private sector is rapidly innovating to meet these demands, cutting energy demand and investing in R&D for clean-energy development and efficiency. For instance, by 2022, Google had already invested $3.5 billion in renewable energy globally and its investments are only accelerating.

Leaders who can strategically harness these investments can help their countries become AI and clean-energy powerhouses: governments that can ensure AI and energy investments go hand in hand can attract not only the infrastructure necessary to fuel the AI-driven economy, but the clean energy required to power its demand.

To do this, governments need to make it easier for companies to innovate and adopt greener AI techniques, including through increasing collaboration between academia and the private sector, increasing R&D investments in green-AI software and hardware, and engaging in international collaboration.


Devorah West, Senior Policy Advisor, Climate & Energy Policy
Devorah West, Senior Policy Advisor, Climate & Energy Policy



Read “Greening AI: A Policy Agenda for the Artificial Intelligence and Energy Revolutions


Subscribe to the TBI newsletter to get insights, research and reports straight to your inbox, as well as the latest TBI updates, including webinars and event news. Just fill out your preferences and we’ll tailor content to you. Sign up now .


Lessons From Singapore

Singapore has set itself apart as a global leader in artificial intelligence, consistently punching above its weight in international rankings of AI readiness and innovation. With the release of its second National AI Strategy in December 2023, the country doubled down on efforts to advance AI and contribute to global AI development – particularly towards use cases for the public good.

Exponential growth in the demand for AI has resulted in a surge in energy demand. Expert analysis indicates AI servers could consume the same amount of energy as some countries by 2027. Supplying this energy poses a serious challenge to those aiming to capitalise on the AI revolution. By prioritising energy efficiency and decarbonisation, Singapore is not only addressing local environmental challenges but also setting a benchmark for global best practices in sustainable AI development.

Digital sustainability is emerging as a key focus of Singapore’s leadership. The country has launched several initiatives aimed at greening AI, such as green software trials and a “Green Data Centre Roadmap”. The green software trials will see industry partners apply carbon-reduction techniques to digital applications to measure their impact in real-world contexts. This will be among the first attempts to understand the cost and energy-efficiency gains of carbon-reduction techniques without sacrificing performance – a huge step forward in creating guidelines for developing greener software. And the Green Data Centre Roadmap, launched last week, aims to chart a sustainable path for the continued growth of data centres in the country, including through the provision of additional green-energy capacity. This builds on earlier efforts to green digital infrastructure, after Singapore released the world’s first set of sustainability standards for data centres in tropical climates.

Central to Singapore’s success in these areas is its commitment to the triple-helix model of innovation, which recognises the crucial interconnection between government, industry and academia to accelerate progress in AI. This approach has been instrumental in driving R&D capacity and rapid translation into deployed solutions, leading to significant advancements in both hardware and software. Singapore’s Sustainable Tropical Data Centre Testbed and green software trials exemplify how these partnerships can produce practical solutions to sustainability challenges.

But Singapore’s leadership extends beyond its borders. The country actively participates in international AI and sustainability initiatives, including the Global Partnership on AI and the Green Software Foundation. By sharing its expertise and collaborating on global standards, Singapore is fostering a cooperative approach to AI governance and digital sustainability that other nations can emulate.

Singapore is paving the way for a sustainable digital future, offering valuable lessons and setting a powerful example for the global community.


Marie Teo, Advisor - Global Government Engagement - Asia
Marie Teo, Advisor, Global Government Engagement, Asia


Read “From Strategy to Synergy: What Can We Learn From Singapore’s AI Journey?


Stay in Touch

要查看或添加评论,请登录

社区洞察

其他会员也浏览了