KEYNOTE: How to Leverage Data to Combat Climate Change 🌳
summary
Claims that AI and machine learning can fill missing company-level carbon data and immediately enable neutralization by buying regulated allowances. Uses a convolutional neural network (CNN) example that maps Chesapeake Bay at 94% accuracy and cites Microsoft reducing building energy by over 20% with ML. Describes building a Python model from 50 employee survey responses with a 90/10 train/test split and extrapolating from ~4,000 CDP reporters to ~14,500 public companies. Neutralization happens via serialized compliance markets where one allowance = one metric ton.
AI-written summary · about
chapters (8)
- 0:00Introduction to climate data challenge
- 1:50Businesses lack actionable climate data
- 6:45Sophia’s grassroots data quest
- 9:15Real-world client success with TXI
- 12:30Machine learning to bridge data gaps
- 16:48Climate Vault’s mission and approach
- 19:19Path forward for climate action
- 21:01Q&A on impact and scale
related talks
22mLet's Terminate Climate Change with AI
24mAI in (climate) action: How to fix carbon markets
29mHow AI and Deep Learning could effort the Climate Change
57mThe Next Decade of Software Is About Climate - What Is the Role of ML?
45mGreen AI: Making Machine Learning Environmentally Sustainable
01h 02mJoint Keynote Address: Scaling AI Sustainably: An Uncharted Territory
28mAI vs Climate Change: Insights from the Cutting Edge
28mData as the new commodity: AI. Sustainability and Financial Innovation