With energy, water and carbon emissions under 5% that of LLMs, GLYNT.AI is the sustainable way to prepare sustainability data

MOUNTAIN VIEW, CA, UNITED STATES, December 5, 2024 /EINPresswire.com/ -- GLYNT.AI Inc., The Sustainability Data Company, announced today that its innovative AI technology leads to lower energy and water use when using AI to prepare data than would be used by LLMs or typical cloud AI tools for the same task. GLYNT.AI released its white paper, “The Sustainable Way to Prepare Sustainability Data” which details how it measured its level of energy, water and emissions, and how the result compares to other AI algorithms used for the same tasks.

GLYNT.AI prepares accurate, audit-ready sustainability data and reporting for investors, customers and regulators by capturing highly varied source files, extracting data using AI, with additional services for enrichment and customization. At the heart of GLYNT.AI’s platform for automated, accurate, audit-ready sustainability data, is the company’s proprietary implementation of “Few Shot” Machine Learning (ML). This innovative algorithm is compute efficient, leading to dramatically lower levels of energy, water and carbon emissions from its services.

“When building our technology platform, GLYNT.AI faced a huge challenge: How to unlock the key data needed for sustainability and climate reporting that is trapped in invoices, utility bills, manifests and other PDFs, CSVs and JSON files.” said GLYNT.AI CEO Martha Amram. “The files are hugely varying and change over time. So we needed a powerful AI system to unlock the data at scale while maintaining accuracy.

“We built a deployment of “Few Shot” ML that uses a library of tiny, tiny training models. Each model can be trained in minutes and needs just three training samples. The model returns data at greater than 98% accuracy rates from the start and so very little retraining is needed. Our hyper-efficient use of computing resources leads to energy and water usage that is less than 5% that of LLMs, and less than 20% of the emissions from standard AI tools from cloud providers for the same task.”

GLYNT.AI documented its calculations and findings in the white paper released today, “The Sustainable Way to Prepare Sustainability Data.” Based on research from Esty, ThoughtCloud, Yale, Meta, Hugging Face, ERPI and LNBL, the white paper details how GLYNT.AI makes comparisons between its ML algorithm and other peer-reviewed findings. The methodology is compliant with the emerging NIST and ISO standards for AI.

“Technology strategy is converging with energy and water markets,” said Amram, “as many electric utilities are projecting double-digit price increases from the rapid growth in data centers and the shift to electrification. Businesses everywhere are looking for strategies that save energy and water, and reduce emissions. This study shows how the choice of algorithm, the choice of third-party services and the data center infrastructure come together, and deliver a menu of reduction options. While the calculations are on the leading edge, and the analysis of resource intensity of various algorithms is in an early stage, the results are clear: GLYNT.AI is hugely efficient. This provides a strong business case and a reduced environmental footprint.”

To learn more, download the white paper at GLYNT.AI

See a product demo at https://glynt.ai/

ABOUT GLYNT.AI
GLYNT.AI is The Sustainability Data Company, and prepares investor-grade sustainability data for finance and sustainability teams around the world. Our accurate, complete and audit-ready data enables reporting, operational efficiencies and access to financial capital. With purpose-built machine learning, GLYNT is the single automated solution for water, waste, energy and emissions data at SOC 1 (SOX) compliance levels. Get the sustainability data job done with confidence. Learn more at https://glynt.ai/.

Martha Amram
GLYNT.AI
+1 650-208-6532
email us here
Visit us on social media:
X
LinkedIn
Instagram
YouTube

Legal Disclaimer:

EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.