by Simon Thompson.
In recent years, rapid advances in machine learning, deep learning, neural networks and large language models have accelerated the development of powerful AI models. Capable of rapidly analysing structured and unstructured data; identifying and recognising patterns; and processing and generating human-like text, audio and video, this has led to the swift adoption of AI in many areas of business and finance, especially in financial services. Use cases involve risk analysis, trading, client and customer support and the automation of disclosure and reporting.
Like any new technology, AI can have both a positive and negative impact on sustainability, and on financial services. It opens new possibilities for financial institutions to address the complex challenges of climate change and sustainability, especially in terms of improving climate and sustainability risk management; and automating and enhancing verification and reporting.
Here are 4 ways in which AI can play a transformative role in sustainable finance:
1. Climate Risk Modelling and Scenario Analysis: AI can process vast amounts of structured and unstructured climate data, facilitating the development of more accurate global and local climate models and more accurately forecasting potential impacts of different climate change scenarios. This helps scientists, policymakers and finance professionals make more informed decisions. AI can also link climate and impact models with lending and investment data, helping financial institutions better understand physical and transition climate risks and align portfolios with sustainability goals. Scenario analysis tools powered by AI can help financial institutions comply with regulatory stress tests and understand potential outcomes under different climate scenarios.

2. Data Analysis and Decision-Making: AI can process and analyze large volumes of ESG data, helping lenders and investors assess the sustainability performance of companies more accurately, and more rapidly. In addition to structured, quantitative data, AI can extract insights from unstructured data, including sustainability reports, news articles, regulatory filings and social media. These can help financial institutions and finance professionals develop a more holistic view of portfolio companies’ sustainability, including the views of a wider range of stakeholders.
3. Detecting Greenwashing: Related to the above, AI can detect inconsistencies and potential greenwashing in corporate and fund sustainability claims by cross-referencing public statements, regulatory filings, and independent audits with news articles, social media, etc. Flagging companies or funds that appear to be misrepresenting their sustainability commitments, ensuring compliance with evolving regulatory requirements and, more broadly, market integrity.
4. Automating Impact Measurement and Reporting: AI can streamline the measurement and reporting of non-financial impacts by automating the tracking of sustainability metrics, such as carbon emissions reductions, energy efficiency improvements, or social impact outcomes. By standardizing and automating impact reporting, AI reduces costs and complexity, enabling smaller organizations to participate in sustainable finance markets. Also reducing the demands on sustainability professionals to produce reports, freeing time for more impactful sustainability activity.
To that AI and finance are genuinely sustainable, however, and as AI continues to develop, we must consider and respond to ethical issues and concerns and ensure the responsible and equitable deployment of AI for the benefit of current and future generations.
Here are 4 potentially negative impacts of AI on sustainability:
1. Energy consumption: AI systems, particularly deep learning models, require significant computational power, which can contribute to increased energy consumption and carbon emissions. The training and operation of AI models may require large data centres that consume substantial amounts of electricity, potentially offsetting the positive environmental impacts where these are not powered by renewable energy.
2. Ethical concerns in decision-making: AI algorithms are trained using historical data, which may contain biases and perpetuate societal inequalities. When used in sustainability and sustainable finance related decision-making processes, such as resource allocation or lending and investment, biased AI systems can perpetuate environmental and social injustices and exacerbate existing inequalities.
3. Data privacy and security: AI relies on large amounts of data to train and make predictions. The collection and storage of sensitive environmental data can raise concerns about privacy and security. Inadequate protection of this data can result in unauthorised access, misuse, or breaches, compromising individual privacy and potentially hindering sustainability initiatives.
4. Job displacement: AI automation has the potential to disrupt many sectors, leading to job losses for certain professions, including in financial services. While AI can create new job opportunities, the transition may be challenging for individuals and communities who lose their jobs due to AI implementation.
Financial institutions and finance professionals need to consider both the opportunities and challenges associated with the use of AI and its impacts, therefore, to ensure a responsible and ethical approach.
Simon Thompson is the author of Green & Sustainable Finance: Principles and Practice (Kogan Page), former Chair of the UK Sustainable Finance Education Charter, and Chief Executive of the Chartered Banker Institute (2007-2024).