We're adding OpenAI to our list of sub-processors. See why and what that means.
What is OpenAI
OpenAI Inc. is an artificial intelligence (AI) research organization. They develop language models, which are advanced AI systems capable of generating human-like text for tasks like writing, answering questions, and more. OpenAI emphasizes the importance of AI safety, ethical usage, and transparency. It works to address challenges such as bias, misinformation, and the responsible deployment of AI.
Why we use OpenAI
We use OpenAI to streamline and automate complex, time-consuming tasks that would otherwise require significant manual effort from our customers. By leveraging advanced AI capabilities, we can instantly calculate product carbon footprints and offer actionable recommendations for reduction initiatives. This not only saves time but also enhances accuracy and scalability, empowering businesses to take immediate, informed action. Ultimately, we believe that integrating OpenAI’s technology is essential for driving rapid carbon reductions, helping to address the global climate challenge with the urgency it demands.
Use of OpenAI
Use cases
Emitwise uses OpenAI in specific parts of the product:
-
Product Carbon Footprint (PCF) Product Classification Model. We use OpenAI Completions to classify bill of material (BOM) line items to a product from our product catalogue, so we can automatically assign emission factors and calculate emissions related to materials. In this case, we send the material's name, material's supplier and material's description to OpenAI
-
Reduction Initiative Chatbot. We use OpenAI embeddings to embed user questions, retrieve relevant answers from a database of public cases studies, and we use OpenAI Completions to generate answers to the user’s questions. In this case, we send the user questions to OpenAI
-
Reduction Initiative Structurer. We use OpenAI Completions to extract content from a file provided by the user and structure it into dedicated fields. In this case, we send the file provided by the user to OpenAI
Data security
Data is sent to and temporarily retained in OpenAI for processing and generating AI output. Only anonymised data is sent to OpenAI, and our agreement prohibits OpenAI to use the data for model training.
OpenAI has been carefully selected and audited to ensure compliance with Emitwise’s standard security practices, including data encryption, access control, and regular security audits. All user interactions are protected under Emitwise’s privacy policy. The use of publicly accessible data ensures compliance with both internal guidelines and external regulatory requirements for ethical AI deployment.
Where Large Language Models (LLM) are used with OpenAI, additional precautions were taken to ensure ethical AI development: the output of the model is reviewed by human experts, the model has a limited output range, and the decisions it makes are only used for the purposes of carbon accounting.
Visit our list of sub-processors and privacy policy.
Visit the OpenAI Security Portal.
OpenAI is optional
If you do not wish to use OpenAI, please reach out at support@emitwise.com and we will take necessary steps to allow you to use our product functionality without the use of AI.