AI: The Underleveraged Asset in the Race to Net Zero

Our fourth annual Destination Net Zero report is an empirical evaluation of where the world’s largest companies stand on their decarbonization journeys. Looking across industries, the world’s largest companies are making progress toward their decarbonization goals and the global push to net zero. Unfortunately, progress has been insufficient to achieve net zero by 2050.

Of the world’s 2000 largest companies (G2000) that disclose sufficient emissions data, over half have reduced absolute emissions and emissions intensity since 2016. Almost two thirds (65%) of the G2000 have net zero commitments that cover at least their operations (Scopes 1 and 2). And of the 21 “decarbonization levers” for which we sought evidence, only five were adopted by more than 80% of companies. The greatest concern is that only 16% of companies are on track to reach net zero by 2050.

We simply must do more and do it faster. AI can help.

In the quest to achieve net zero targets, AI is an underleveraged asset. As highlighted in our recent collaboration with the United Nations Global Compact, AI can significantly accelerate net zero performance by optimizing energy use, improving operational efficiency, and enabling real-time decision-making. It can predict emissions hotspots and unlock new business models, demonstrating AI's potential to transform corporate decarbonization efforts. In our new Destination Net Zero report, we feature pioneering companies that are already using AI to advance their net zero goals. Saudi Aramco is using AI to increase operational efficiency and reduce carbon emissions, and Microsoft is adjusting energy usage in real time with smart building technologies. Additionally, AI can enhance ESG reporting. With regulatory bodies worldwide implementing increasingly stringent sustainability disclosure and performance mandates, companies with faster access to ESG data, from across the extended value chain, are positioned to move from mere compliance to competitive advantage.

The opportunity to significantly accelerate decarbonization with AI, and particularly gen AI, does come with notable cautions. Monitoring AI’s impact should be addressed as part of an integrated responsible AI framework that is able to quantify the emissions associated with AI to ensure that the sustainability benefits aren’t outweighed by its carbon footprint. Ultimately, the overall emissions impact of AI will depend on how effectively these applications are deployed and on reducing AI’s energy footprint through responsible technology choices. Careful consideration of energy usage, infrastructure and operational efficiency will be key.

By embedding sustainability principles— functionality and responsibility — into AI adoption, companies can unlock significant decarbonization potential. AI can help drive better, faster outcomes in:

1)         Planning Companies with targets and transition plans typically reduce emissions faster than those with targets alone, yet over half of the G2000 do not have a climate transition plan in place. They need to take action. Organizations should aim to develop transition plans to reach their targets, incorporating near-term actions and milestones and ensuring validation by credible external bodies. AI can help with this. Real-time monitoring and predictive analytics are invaluable for setting and managing transition strategies. AI can also improve the way plans and performance are communicated. Accenture’s ESG-specific specialized language model (SLM), for example, helps companies automatically structure and generate report components based on key metrics, achievements and regulatory requirements. This helps enable the rapid, consistent production of high-quality, accurate ESG disclosures across industries.

2)         Optimizing AI can also assist companies in prioritizing decarbonization levers that align with business models and address specific emissions hotspots. Companies should implement foundational levers such as improving building efficiency. Retrofitting with energy-efficient materials, or switching to low-carbon heating, are measures that are likely to be relevant to all companies. With AI, specific levers like this can be augmented: building occupation can be monitored, and energy use (e.g., lighting and heating) can be adjusted in real-time to match demand. Equally, AI becomes invaluable when companies seek to make sense of their data to identify the most efficient ways to reduce their emissions.

3)         Pushing the frontier  Beyond using AI to support with planning and optimizing existing efforts, companies can use it to find and pursue new ways to decarbonize. This can mean using AI to identify different materials or ingredients to replace less sustainable default choices. Unilever is for example using AI to detect possible alternative ingredients and remove less sustainable ingredients from its products, without compromising on quality. AI can also help design new, low-carbon products to meet a niche market need.

AI is already a powerful business tool, boosting productivity and innovation for people and companies alike. With its ability to enable real-time decision-making, predict emissions hotspots and unlock new business models, AI has the potential to accelerate corporate decarbonization efforts. It could be the game changer we so desperately need.