Data Science Institute Director Discusses How Emerging Technologies Affect the Environment: What Impact Does AI-Machine Learning Have on the Environment?

Does applying an algorithm use a lot of energy? Does training a neural network on a data center leave a significant carbon footprint? And what impact do Al and machine-learning models have on the environment?

In this Market Watch video, Jeannette M. Wing explores these questions.

"In computer science, for decades we have measured the performance of algorithms along two dimensions, time and space," says Wing, Avanessians Director of the Data Science Institute and Professor of Computer Science at Columbia, "How much time does it take the algorithm to compute the final answer. How much space or memory does the algorithm take in computing the answer. Now, with energy being so important in everyone's minds, the computer science community is paying more attention to a third measure of the performance of an algorithm and that is energy usage."

In the video, Wing explains that about 2 percent of total carbon emissions is generated by the Information and Communications Technology (ICT) sector. Nonetheless companies, especially smaller ones, should consider using cloud computing services, which are more energy efficient than data centers run by smaller firms, she says.  

"We are really at the tip of the iceberg in terms of the use of Al and machine-learning models," says Wing, "so the days of studying this question are just beginning."  

–Robert Florida

550 West 120th Street, Northwest Corner Building, Suite 1401, New York, N.Y. 10027    212.854.5660
©2019 Columbia University