Inspiration

Our team is working on the development of a more comprehensive tool - the RegenX Investor Scenario Tool (https://conservationx.com/project/id/1600/regenxscenariotool) - which uses AI and science-based scenarios to put ESG data in context for investors. This tool will comprise cross-sector data analysis, degenerative versus regenerative scenarios, and supporting narratives to guide investors towards regenerative investments. It will serve investors by potentially reducing risk in their portfolios and surfacing new opportunities.

Given the importance of CDP data in the wider ESG landscape, we chose to enter this hackathon to test some of the ideas and principles in our tool development process, in the spirit of collaboration and sharing with CDP, a critical stakeholder in the climate-aligned investment space.

What it does

Our solution to CDP’s Challenge #1 seeks to indicate where silos can be taken down to increase collaboration, and to demonstrate the direction of travel in terms of regeneration for North American companies and cities.

We used AI to look for regenerative indications, based on the Future-Fit Business Benchmark's goals and technical information. We chose this as a proxy for the best compilation of science-based and context-based targets.

How we built it

For this iteration we focused on water data as a pilot but the approach can be applied to any element of a business and its sustainability initiatives. Our intention is to broaden the scope of the data categories - e.g. to climate and forests - and to increase the sophistication of the AI discovery process.

Within the array of survey questions we prioritized CDP Corporate water and CDP City responses that relate to strategy and targets because we feel these best describe how a company / city sees its purpose, and because these free-form fields are the hardest to work with efficiently so the AI tool can accelerate the analysis.

Challenges we ran into

We needed to find a way to reconcile disparate data with a clear understanding of the characteristics of regeneration. This required several iterations of “training” the AI tool to recognize certain words and phrases.

We also encountered a challenge with the fundamentals of the data. These disclosures are voluntary and there is no clear standard for much of the information - especially the responses that could indicate how committed a company is to regeneration.

Accomplishments that we're proud of

Our regeneration benchmarks draw on objective, science-based targets. This represents a shift away from relative comparison towards context-based ESG analysis, which is a critical distinction for generating net-positive impact.

Our simple, intuitive visualizations help users find opportunities for bioregional cooperation between cities and corporations.

What we learned

We learned that it is possible to use AI to discover seeds of regeneration in current sustainability frameworks.

What's next for The RegenX factor

We will continue to refine the AI tool to accurately identify companies on the path to regeneration. We will also broaden the tool’s discovery process beyond water risk to include climate and forests, and to seek opportunity-related activities as well as risks.

We will also be engaging directly with potential users of this tool - specifically long term institutional investors - to accelerate the shift of capital into regenerative channels.

Built With

Share this project:

Updates