AI and the Ethics of Energy Efficiency

I was accepted to a year-long Fellowship hosted by the Markkula Center for Applied Ethics at Santa Clara University that explored the intersection of AI, ethics, and the environment out of 100 applicants. The Fellowship was a self-guided research program with the goal of publishing an article of a topic of my choice under the guidance of Dr. Brian Green. Since taking AP Environmental Science in high school, sustainability has remained a passion of mine that I try to incorporate into both my daily life and my work. My passion for designing a sustainable future combined with my passion to innovate through technological design, specifically in the realm of AI, led me to explore whether the use of AI produced a net negative or positive energy consumption. My interest was sparked after taking a course in the Ethics of AI and exploring how rapidly AI was becoming used in modern systems, and more specifically after reading an article about Google's DeepMind use in data centers used for cooling. After months of research through the fall of 2019, I discerned that despite the energy consumed to train AI, AI still produces a net negative energy consumption due to the energy it saves through its implementation. This research also continued to reinforce the pressing need to evaluate the context technology is implemented in, its long-term effects, and how humans interact with technology to make it run as efficiently and positively as possible. My article was published at the end of the academic year and I continued to research the use of AI in graduate research courses.