The future of climate risk reporting: design-led approaches to improving integrity
The Challenge
Project Duration : January 2024 – June 2024 (pilot phase)
Green plantation
Climate risk disclosures are documents published by companies and other institutions to provide information about the environmental impact of their activities along with their exposure to climate risks. The goal of sharing climate-related disclosures (and other forms of environmental reporting and regulation) is to: (i) ensure that the impacts of climate change become a core consideration in all business and investment decisions, and (ii) provide a source of accountability and transparency that companies and institutions are disclosing and mitigating risks associated with climate change and environmental impacts.
Currently, there is no efficient way of assessing the accuracy of the claims made by companies and institutions in their environmental reporting. As providing these disclosures becomes a mandatory obligation, the challenges for companies and institutions to not only comply with this requirement but also provide a rigorous evidence base for their claims are significant.
Climate risk disclosures are one example where companies and institutions need support to improve the integrity of their environmental reporting and provide assurances for their plans in pursuit of sustainability targets. Similar issues exist in ESG reporting, (forthcoming) nature-based disclosures, and broader sustainability credentialling and reporting: they are each rapidly evolving spaces.
In the face of these challenges, how might we reconsider the data and information underpinning climate disclosures (and the work required to generate them), to assure robust claim making and integrity of action?
Our response
CSIRO’s Responsible Innovation Future Science Platform, and Data61 and Environment Business Units are collaborating on this pilot scale project to explore the intersection points climate risk disclosure reporting, environmental and financial value realisation, and an unprecedented capability in Generative Artificial Intelligence.
In this project, the research team leverages each other’s strengths across diverse areas of behavioural, environmental and computer sciences. Combined with their insights working with industry stakeholders, the team is exploring ways of promoting companies’ ambitions for reducing emissions, preventing damage to natural ecosystems, and pursuing bold sustainability targets. By combining new generative AI approaches with ‘Warm Data Lab1’ methods, the project aims to recontextualise these complex systems of reporting valuation and risk. Specifically, the generative AI component explores how a Large Language Model might assist with report generation and/or disclosure assessment while the Warm Data Lab offers an additional dimension to what can be learned from quantitative data (i.e. ‘cold data’).
Given the rapid rate of development and adoption of AI can affect legitimacy concerns (e.g. greenwashing) and regulatory considerations (e.g. regulation of AI and reporting requirements), the team will explore how environmental reporting information could be recontextualised for elements of legitimacy, accountability, and ambition so that new investment opportunities may be realised. This novel combination of methods is built around a strong commitment to human-centred design and design science2.
Impact
The aim of this research is to investigate and test what methods and criteria can be employed to effectively assess the veracity of claims made in climate risk disclosures, and how these approaches can be refined to ensure responsible, transparent, and sufficiently ambitious corporate reporting on climate-related issues.
This six-month pilot project is a design-oriented ‘discovery’ research project focused on a future horizon. It includes an exploration of current industry and technology trends so that the opportunity space can be formatively understood. The immediate goal is to propose a proof-of-concept solution informed by a range of expertise and ambition that will be further developed through a larger scale research investment.
The longer-term ambition is to consider possible futures where new capabilities could catalyse greater environmental accountability and investment.
Team
A deliberate cross-BU team, covering a range of scientific disciplines and backgrounds
Project Leader, Cara Stitzlein, Sam Nicol, Sonja Dominik, David Evans, Chantelle Goetz, Maciej Rybinski, Scott Spillias, Simon Fielke
Related Links
1 For information on Warm Data Labs, see https://humanrights.gov.au/sites/default/files/2020-07/22_-_phoensight_1.pdf and https://warmdatalab.net/
2 The project operates in the ‘Discovery-Define’ phase of a Double Diamond design framework, see Framework for Innovation – Design Council