AI Tool Reveals Old Scope 3 Models Overstate Emissions by 2,480%

New findings by UK-based carbon data firm Carbon Responsible have found that old methods used to estimate Scope 3 emissions are capable of overestimating actual emissions by up to 2,480%. These old models are still the core of a number of reporting frameworks, including the Greenhouse Gas (GHG) Protocol and the Partnership for Carbon Accounting Financials (PCAF), despite tighter emissions reporting standards in both Europe and the US.

Scope 3 emissions are indirect emissions along a company's value chain, including supply chains, product use, employee commuting, business travel, and waste disposal. The emissions can represent more than 80% of a firm's total carbon footprint. The research, contrasting traditional estimation techniques with real emissions data of FTSE 100 firms in 2023, looked back at the enormous discrepancies depending on outdated models.

Most traditional methods, such as Environmentally Extended Input-Output (EEIO) modeling, rely chiefly on historical spending estimates. These in turn use outdated and aggregated data sets, which do not reflect recent supply chain arrangements or altering market trends. Despite this, they are used across the board due to their simplicity and adaptability with current disclosure standards.

But as regulators move towards verifiable and mandatory reporting frameworks, the shortfall of old estimation tools is becoming more apparent. According to the European Union's Corporate Sustainability Reporting Directive (CSRD) and other tough regulatory guidelines in the United States, companies are soon going to be required to report auditable and traceable emissions data. Such provisions do not make use of the estimation-based procedures that were once tolerable.

In order to solve this problem, Carbon Responsible built an artificial intelligence-powered platform named Ada. The platform is employed for giving very precise Scope 3 emissions calculations based on a database that has over 14,000 verified company listings. These listings are regularly updated and do not have any old data included to maintain it current in real-time. Ada uses machine learning to process and distill this data so that it can produce 30 times more precise results than traditional EEIO techniques.

According to Carbon Responsible, the accuracy can be increased by 97% and estimation error can be reduced by up to 80% to transform what started as an approximate estimate into investment-grade data. These are all enhancements making the platform better suited to meet the demands of today's global regulations and business ESG needs. As opposed to legacy models that rely on generalized industry data, Ada provides bespoke reports based on verified activity data, enhancing the transparency and defensibility of the process.

The report also suggests that the long-term application of classical emissions estimation models could place the businesses at reputational and regulatory risks. As audit requirements keep on rising and the genuineness of reported data becomes an area of concern, investors, regulators, and ESG rating agencies are putting more value on credible and accurate reporting. Companies that fail to upgrade their emissions monitoring systems might not be able to comply with regulation needs and get sustainability-focused investment.

Another implication of this shift is the increased importance of mainstreaming real-time data analysis in sustainability reporting practice. The use of AI and machine learning makes it possible for business companies to process vast quantities of complex data efficiently, generating insights that inform strategic decisions and goal identification. Technologies like Ada can assist companies in identifying hotspots for emissions, evaluating supplier performance, and designing targeted intervention for emissions reduction.

The move towards more precise Scope 3 emissions reporting is in line with industry trends overall. ESG performance is increasingly being linked to financial performance, and as stakeholders demand greater accountability, integrity of data must be a priority. Platforms that can provide very high-quality emissions data not only facilitate compliance but also facilitate trust establishment and competitive differentiation in an oversaturated market.

Carbon Responsible's technology strategy represents a departure from generic modelling to verifiable, specific data. This trend will have implications for corporate emissions reporting, validation, and auditing in the years to come. As companies prepare to meet future CSRD and other comparable regulatory requirements, technology-based solutions like Ada will be favorite options for sustainable performance and value creation in the long term.

Because Scope 3 emissions account for the lion's share of corporate emissions, the findings of this study have far-reaching implications for ESG reporting standards, corporate compliance, and climate accountability. As regulators redefine acceptable carbon reporting standards, AI platforms are emerging as a critical piece of the solution, bringing transparency, accuracy, and scalability to accounting for emissions.

Source/Credits:
Funds Europe