However, a significant gap remains between intentions and actions. Businesses often set ambitious targets but struggle with creating paths to achieve them. Also, sustainability is often separated from other key corporate functions like strategy and innovation. In addition, there are challenges related to incomplete, inaccurate or fragmented sustainability data.
Could generative AI (gen AI) help drive sustainability in organisations? It can handle large, unstructured data sets and pull actionable insights from sources like reports, news, and public databases.
Also, tools like chatbots and AI-powered assistants could make sustainability insights accessible across an organisation, encouraging everyone to get involved.
Could the benefits of gen AI ultimately outweigh its environmental costs? Let’s dive into some concrete examples of how gen AI could enhance sustainability efforts within organisations.
1. Sustainability strategy and initiatives
While 90% of executives think sustainability is important, only 60% of companies have a sustainability strategy. Businesses often set ambitious goals but face challenges in defining clear steps to reach them.
Gen AI could support organisations in developing and refining their sustainability strategy. By analysing data like past initiatives, regulatory shifts, market trends, and consumer preferences, it can identify key challenges, risks, and opportunities for innovation.
For example, organisations could use a gen AI-powered assistant to explore and evaluate new circular business opportunities. It could provide insights into the potential financial benefits and environmental impacts of different circular strategies.
Or they might use it to find the best ways to reduce Scope 3 emissions – those that come from activities outside of their direct control, like emissions produced by suppliers or during transportation. These emissions are often hard to measure because supply chains can be complex, with lots of different companies involved. By analysing data like material types and transportation methods, gen AI can estimate emissions of specific activities.
For example, a SaaS platform CO2 AI enables business leaders to develop an accurate estimate of their organisations’ Scope 1, 2, and 3 emissions down to the product level. It also helps them to model and evaluate emissions reduction opportunities.
Gen AI can also assist organisations in defining relevant sustainability goals and KPIs, and tracking progress against them. Imagine an organisation asking their AI assistant, “What KPIs should we establish to measure our progress toward a 50% reduction in greenhouse gas emissions by 2030?” The AI could explore industry benchmarks and best practices to suggest metrics like the percentage of renewable energy used.
With many sustainability initiatives to choose from, it can be tough to know where to focus your efforts. Gen AI can simulate different scenarios and prioritise initiatives by evaluating them against key metrics, like return on investment and environmental impact.
For instance, an organisation might use a gen AI-powered assistant to ask, “What would be the projected ROI and carbon reduction of adopting a circular economy model compared to investing in energy-efficient technologies in our facilities?” By analysing these scenarios, the AI can provide insights on which initiative offers the greatest potential impact.
Finally, the sustainability landscape is always evolving, so strategies should adjust based on real-world insights. As new data comes in—whether it’s regulatory updates, feedback from initiatives, or shifts in technology—AI can continuously learn and adapt your strategy.
2. Sustainable service and product innovation
When it comes to creating new products or services, gen AI can act as a useful brainstorming partner. For example, a study of over 750 BCG consultants revealed that using GenAI (like OpenAI’s GPT-4) for creative product innovation led to a 40% performance boost.
Gen AI can sift through large volumes of data to uncover insights that aren’t immediately obvious. By examining consumer trends, preferences, and feedback, it can generate ideas for services or products that align with customer needs. However, it’s good to keep in mind that AI can’t fully replace the nuanced understanding gained from engaging with stakeholders. These interactions can reveal hidden needs and deeper motivations.
Gen AI allows teams to rapidly produce multiple iterations of service or product concepts, which can then be refined and developed further in collaboration with real stakeholders.
For example, Nestlé is using gen AI in sustainable product innovation. They are using a tool that is taking in inputs from more than 20 Nestlé USA brands and analysing real-time market trends to suggest creative product concepts. In their early efforts, they have seen the tool accelerate the product ideation process from six months to six weeks.
Also, large language models can speed up innovation by analysing research papers and patent applications, quickly uncovering new ideas and highlighting gaps in existing knowledge.
3. Sustainable supply chain management
Gen AI can help identify inefficiencies and emission hotspots in the supply chain, and recommend actions based on these insights. Its ability to draw insights from unstructured sources—such as news, reports, and public databases—offers an advantage over “traditional AI”, which is better suited for structured data.
For example, gen AI can simulate supply chain scenarios, considering evolving inputs such as regulatory changes, market trends, climate patterns or shifts in resource availability. For instance, it could model how potential regulatory changes might impact compliance needs.
When it comes to sourcing, gen AI can assist in identifying the most sustainable vendors, ensuring each link in the supply chain complies with local regulations and has minimal environmental impact. Gen AI can tap into unstructured sources, such as sustainability reports, supplier websites, news articles, or supplier reviews, to get a fuller picture of a vendor’s environmental and social impact.
Gen AI can also enhance supply chain sustainability by optimising logistics, minimising fuel consumption and emissions. By analysing data from various sources, gen AI can identify the most efficient routes and methods for transportation. For example, it could scan for real-time traffic updates or weather alerts. Based on this information, it could suggest alternative routes, transport modes or delivery schedules.
Furthermore, gen AI can predict demand more accurately, reducing overproduction and excess inventory. Besides using past sales data and seasonal trends, it can access unstructured data like social media trends, economic news, and competitor activities, and refine predictions in real-time.
Also, Gen AI can boost predictive maintenance by identifying patterns that signal potential equipment issues. Besides monitoring real-time sensor data, it could simulate machine performance under various conditions. By integrating diverse data sources—like maintenance logs, environmental conditions, and usage patterns—it could build a more accurate maintenance schedule, reducing unnecessary repairs.
For example, Siemens is releasing a new generative artificial intelligence (AI) functionality into its predictive maintenance solution. Using a conversational user interface, manufacturers can take proactive actions easily, saving both time and resources.
4. Sustainability reporting and communication
With regulations like the Corporate Sustainability Reporting Directive (CSRD) demanding clearer disclosures, effective sustainability reporting is essential for compliance. However, many organisations struggle with scattered, complex and incomplete data, resulting in inconsistent information.
Gen AI (GenAI) can enhance ESG reporting by analysing large datasets and processing unstructured information from various sources. For example, it can process data from corporate reports, supplier practices, regulations, customer feedback, and logistics operations.
By bringing this information together, gen AI can generate reports that not only meet regulatory standards but also tell a compelling story about your organisation’s impact on the planet and people. By presenting data in a user-friendly format, such as engaging narratives and visualisations, gen AI can facilitate transparent communication with investors, regulators and customers.
For example, Schneider Electric introduced an AI tool designed to help businesses query their sustainability reporting data. It equips business leaders and sustainability teams with enhanced data analysis, visualisation, decision support, and performance optimisation.
Also, EnerSys, an industrial battery and energy storage company, is using ChatGPT Enterprise to analyse sustainability data, such as Scope 1 and 2 emissions and waste metrics. They also use it to answer customer surveys and inquiries by uploading sustainability reports and internal policies, though human review remains essential for accuracy.
Tackling AI challenges and getting started
Finally, it’s essential to acknowledge and address the challenges of implementing AI sustainably.
The use of gen artificial intelligence (Gen-AI) is expanding quickly, with significant environmental concerns. Training and running these models demand substantial energy, adding to global GHG emissions. Generative AI systems might already use around 33 times more energy to complete a task than task-specific software would.
On the upside, advancements in technology could reduce AI’s energy demands. AI providers are working to improve energy efficiency and incorporate clean energy sources. Also, AI itself might mitigate emissions by up to 10%. For example, AI applications can help identify decarbonisation opportunities, and optimise production, energy use and transportation.
The key question is whether AI’s benefits can outweigh its environmental costs. More collaboration across different stakeholders and transparency is needed for a balanced assessment. For example, regulators including the European Parliament are starting to require AI systems to log their energy use.
Additionally, training gen AI on inaccurate or biased data can result in misleading outputs, eroding trust and credibility. Gen AI can also produce “hallucinations”- confident but incorrect responses.
Addressing these issues is essential for unlocking the full potential of AI in your organisation. To navigate these challenges, strong human oversight and robust data governance practices are crucial. This includes monitoring AI outputs for accuracy and updating models with reliable data.
While there are several challenges to consider, experimenting with gen AI for sustainability is likely well worth the effort. Are you ready to leverage gen AI as part of your sustainability efforts?