Green Honesty

Inspiration

In the face of the ongoing climate crisis, consumers are leaning towards eco-friendly companies. However, the rise of greenwashing, where companies make deceptive claims about their environmental impact, has made it challenging for consumers to differentiate between genuine and misleading claims. Green Honesty is inspired by the need to ensure companies are truthful about their sustainability efforts.

What it does

Green Honesty evaluates companies' green claims against set criteria, providing an 'Risk Score' that indicates the likelihood of the company's claims being truthful. It references criteria like transparency, evidence-based claims, and more, helping consumers make informed decisions.

How we built it

Using a gold-standard NLP toolkit (BeautifulSoup + Python), when a user enters a company url to evaluate their ERG claims, our app automatically scrapes company websites, cleans and preprocesses the data to get the core of the company's ESG claims. We do this by ensuring the data is coming directly from the company source (the claims they promise on their websites as well as in their annual/semi-annual sustainability reports).

After applying traditional NLP methodology (i.e. company data is scraped, cleaned and ready for processing), we then feed the text through an integrated open-AI model component. This component is based on it's parent multi-giga-model (chat GPT 4.0), but is a derivative of this model in the sense that we provide a concise prompt to the open-AI model, which explicitly spells out what criteria we would like the model to evaluate companies claims on. This in turn gives a scoring of how likely their claims are "true" or "green-washed".

Challenges we ran into

Differentiating between genuine and well-intentioned claims and greenwashing can be subjective. Creating an algorithm that accounts for nuances and provides fair evaluations was challenging. Additionally, we had many paths the vision for this service could take, and narrowing down explicit users and respective use-cases was difficult given the time-constraints of the Hackathon.

Accomplishments that we're proud of

Green Honesty successfully aids VCs, listed companies, regulatory bodies, certification agencies & consumers in identifying genuine green claims, leading to a shift towards truly sustainable consumer choices and holding companies accountable.

What we learned

The complexity of green claims varies across industries. A one-size-fits-all approach might not be the most effective. Instead, a sector-specific analysis offers more accurate evaluations.

What's next for Green Honesty

Green Honesty aims to expand its database its web crawlers and data scrapers, include user-generated reviews, and collaborate with environmental bodies to continually refine its evaluation criteria. We also plan to create partnerships with companies to certify and promote their genuine sustainability efforts.

Built With

• ai
• cloudfare
• natural-language-processing
• python
• suprabase