Demos
Working with Climate Action Veteran Natural Capital Partners, John Snow Labs Minimizes the Environmental Impact Associated with Building Large Language Models
John Snow Labs, the AI for healthcare company providing state-of-the-art medical language models, announces today its CarbonNeutral® company certification for 2024. This certification was done by Natural Capital Partners, an independent global leader in sustainability solutions. John Snow Labs has been carbon neutral throughout its entire history. This effort reinforces the company’s commitment to ethical AI and environmental stewardship by minimizing and offsetting its carbon footprint.
This certification is governed by the internationally recognized The CarbonNeutral Protocol. It requires the team to rigorously measure, reduce, and offset its greenhouse gas emissions.
Achieving this status reflects John Snow Labs’ ongoing engineering, scientific, and operational efforts to minimize the environmental impact of AI technologies. This includes the energy demands of training advanced Large Language Models (LLMs), Natural Language Processing (NLP) models, Visual Document Understanding (VDU) models, and healthcare AI solutions.
“The rapid evolution of Artificial Intelligence comes with immense potential — and responsibility,” said David Talby, CTO, John Snow Labs. “At John Snow Labs, sustainability is a key engineering requirement in everything we build, just like security or scalability. Maintaining our CarbonNeutral certification in 2024 underscores our dedication to operating ethically while addressing the environmental challenges of the AI industry.”
AI Sustainability in Action
John Snow Labs has implemented numerous measures to reduce its carbon footprint, including:
- Optimizing LLM Training and Fine-Tuning: Inventing new optimizations to reduce the compute requirements of creating or fine-tuning LLMs.
- Training “Small” Task–Specific Language Models: Small models can not only be more accurate in specific tasks, but also faster and cheaper to run – requiring fewer energy-consuming compute resources.
- Composing multi-task AI pipelines: For large-scale data analysis projects, engineering one pipeline that processes data once across multiple tasks can be a simpler, cheaper, and more sustainable solution.
- Optimizing Compute Workloads: Reducing the energy intensity of AI model training and inference This involves substantial engineering work with partners like AWS, Azure, Nvidia, Intel, Databricks, Snowflake, and Oracle.
- Supporting Renewable Energy: Offsetting unavoidable emissions through certified renewable energy projects that combat climate change while promoting sustainable development.
- Minimizing Travel: Being a fully remote company reduces real-estate and travel related emissions. Leveraging remote collaboration and virtual events further reduces the environmental impact of business operations.
The Need for Carbon Neutrality in AI
Training large-scale AI models can result in significant carbon emissions, with some models emitting more than 600,000 pounds of CO₂ during development. This reality makes it imperative for AI companies to prioritize sustainability. As a leader in healthcare AI, John Snow Labs recognizes its role in setting an industry example by embedding sustainability into its core mission.
By achieving and maintaining its CarbonNeutral certification, John Snow Labs not only minimizes its own environmental impact but also inspires other organizations in the AI and healthcare sectors to adopt more sustainable practices.
To learn more about John Snow Labs, visit https://www.johnsnowlabs.com/. To learn more about how companies are achieving carbon neutral status, visit https://www.carbonneutral.com/.
