Sasha Luccioni: Connecting the Dots Between AI's Environmental and Social Impacts
The Gradient: Perspectives on AI
00:00
Mitigating Carbon Footprint in AI Experimentation
The chapter focuses on the environmental impact of AI experiments, discussing the carbon footprint of uninformative trials and the need for more mindful experiment design. It highlights the inefficiencies of grid search, the importance of sharing failed experiments, and the challenges of reproducibility in AI research. The conversation delves into the energy consumption of large models and explores ways to develop more power-efficient architectures.
Transcript
Play full episode
