This is an excerpt from 'The Future of ESGTech 2024' report.
As the world plummets further into climate disaster and ecological crisis, every organisation has an obligation to actively work towards becoming more a sustainable and ethical institution.
Within the financial services sector, corporations have an influence on how sustainability is being addressed, to what extent they are working towards net zero targets and decarbonisation, and how new innovations can drive sustainable action. This power
cannot be taken lightly, as more and more consumers and regulators are demanding that companies be held accountable for their environmental impact.
As companies undergo digital transformation and embrace cloud platforms which dial-in sustainable and ESG efforts to the frameworks of a business, artificial intelligence (AI) is the next grand frontier for which financial institutions are embarking on –
and sustainability is an unavoidable path on this journey.
As AI has become a key aspect in new innovations across many industries including financial services, new questions arise: how we can leverage AI to form more sustainable decisions and translate them into action? How do we hold ourselves accountable and
use this new technology ethically and strategically to cultivate the best outcome for our planet and our surroundings? The following chapter will examine AI’s role in the sustainable finance industry and the complexities that arise with AI integration and
an abundance of data.
How can AI be used as a problem-solving tool?
Greenwashing is a prevalent issue in the financial services industries, with banking giants such as HSBC, Citi, and JP Morgan Chase having faced greenwashing accusations in the attempt to appear sustainable and ESGoriented while not putting any real effort
or action beyond a statement. With many ESG considerations, companies self-report, which is proves problematic as there is a risk that they omit the negative aspects of their operations and highlight what they are doing right.
Additionally, sustainable reporting is complex and requires a multitude of details – from scopes 1 to 3 of carbon emissions, environmental impact, energy consumption, and more – often companies are not aware of the correct reporting criteria. These situations
can generate inaccurate reports, making it more difficult for stakeholders and regulators to make informed decisions. AI can solve a major part of the reporting issue in ensuring transparency by bringing together various complex datasets within a company and
form a solid and accurate report.
A 2022 paper by the European Capital Markets Institute outlined that AI could be used in textual analysis to identify controversies on ESG conditions. Tools such as Natural Language Processing (NLP) through software such as RepRisk and Truvalue Labs can
monitor a variety of sources to screen companies’ ESG policies, allowing regulators and governmental bodies to hold them accountable.
Organisations such as the Task Force on Climate-Related Financial Disclosures (TCFD) used machine learning in annual AI Reviews of over one thousand global firms to identify disclosures concerning five different types of climate-related risks in company
annual reports. The use of AI is expanding in the regulatory space to assess risk and compliance with data.
Leveraging satellite and geospatial data with AI to monitor nature-related risk
Satellite, sensor, and geospatial data has also been on the rise in reporting on carbon emissions and environmental impact on biodiversity. These present a new form of data being introduced to the space, monitoring pollution, groundwater quality, deforestation,
waste, and more through geographical coverage. The key advantages of geospatial data are that it is high resolution and is difficult to manipulate.
Laimonas Noreika, CEO and co-founder of HeavyFinance, commented: “AI plays a pivotal role in sustainable finance, particularly when integrating satellite-based geospatial data into city design. The convergence of AI and geospatial data offers transformative
potential in understanding and advancing sustainable practices in urban environments.”
Jon Trask, CEO of Dimitra, a blockchain-based enterprise system for AgTech, explained how geospatial satellite technology and AI is being used to support sustainable farming:
“We use geospatial imagery gathered by drones. We ran a project earlier this year to use drones to analyse corn in Papua New Guinea and essentially built an AI system that will identify where the pests are on the corn so that farmers can precision spray
versus using pesticide completely across the field. In this area, this farming organisation has 11,000 hectares and lost almost 50% of their crop in 2019 due to fall armyworm, and fall armyworm wasn’t an issue before that. Farmers had to change their farming
practices to work with their environment and modify their strategies. Farmers always need to make decisions regarding pests and sustainability and create that balance. So, through our app we help farmers with recommendations and making decisions around sustainability.”
Trask explained that in Brazil, Dimitra’s technology uses geospatial data to measure crop health and report on moisture, nutrients, chlorophyll, and with making carbon estimates. Gathering a wide net of data, Dimitra can inform farmers on how to adjust their
strategies according to what their environment needs, providing farmers with easily-implementable actions so they can farm as sustainably as possible.
Similarly, geospatial data can be used to map out widespread spaces for the objective of sustainable urban planning. By being able to detect specific the state of the landscape, geospatial data can be used to better understand how to integrate architecture
with the needs of the land, and create a sustainable city.
What are the ethical concerns with AI and maintaining transparency?
AI methods can be difficult to discern and can be manipulated depending on the methodologies of collection. There is not a standard form of reporting ESG ratings as yet, therefore when analysing this data there is a risk that it can be manipulated.
On the advent of AI, Noreika stated: “The vast volumes of asynchronous data from shareholder disclosures, satellites, and social media present both challenges and opportunities. AI enables the processing of this data at an unprecedented scale and speed,
uncovering valuable insights regarding environmental, social, and governance factors that impact investments. By analysing this data, we can gauge the sustainability performance of companies, identify trends, and inform investment decisions effectively.
“However, inferring meaning and making decisions based on this data requires careful consideration. AI-driven algorithms must be transparent, interpretable, and unbiased to ensure equitable decision-making. Furthermore, the quality and reliability of data
sources must be rigorously assessed to avoid misinterpretation or greenwashing.”
While AI technology could open up new avenues for growth and development in the sector, especially when it comes to monitoring sustainable efforts and holding corporations accountable for their emissions and environmental impact, it is critical that regulation
keeps AI in check and that data sources can be traced back, to avoid AI becoming a black box technology.
Noreika also commented on ethical concerns of using AI capabilities in the sector: “It is possible for AI to create issues, such as an over-reliance on automation, which could lead to a lack of human oversight and accountability. Ethical concerns regarding
privacy and data security must be addressed to maintain trust in AI-driven sustainable finance practices. It is crucial to strike a balance between the opportunities AI presents and the ethical and transparency challenges it poses, ensuring that sustainable
finance remains responsible and equitable.”
As datasets vary based on the structure of their collection, biases can emerge within the data. To ensure that new AI technologies are without bias, there needs to be a collaborative effort in generating AI and producing explainable AI platforms that maintain
AI offers numerous benefits and multiple paths towards growth, efficiency, and action within the sustainable finance sector. However, there are downfalls to the endless data available and a likelihood of bias within data collection. Moving forward, financial
institutions must consider the risks of integrating AI capabilities into their sustainable finance agendas as well as the reward, and determine what can be done to mitigate the risks wrought by AI technology.