LAWRENCE — The discussion surrounding the environmental impact of artificial intelligence has intensified with its advancement. A recent study revealed that AI emits significantly less carbon compared to humans when it comes to writing and illustrating tasks. Despite this efficiency, the study’s authors emphasize that AI should not be seen as a complete replacement for human writers and illustrators.
Co-authored by Andrew Torrance, the Paul E. Wilson Distinguished Professor of Law at KU, the study compared various established systems like ChatGPT, Bloom AI, and DALL-E2 in completing writing and illustrating tasks against human counterparts.
Similar to the debates around cryptocurrency, AI has been under scrutiny for its energy consumption and environmental implications. While human emissions and environmental impact have been extensively researched, direct comparisons between AI and human carbon footprints have been limited. The study conducted by the authors revealed that AI systems emit significantly less CO2e (carbon dioxide equivalent) per page of text and image compared to human writers and illustrators.
The study, published in the journal Nature and co-written with Bill Tomlinson, Rebecca Black, and Donald Patterson from the University of California-Irvine, aimed to provide data-driven insights into the efficiency of AI in reducing environmental impact.
By calculating the carbon footprint of a person engaged in writing tasks, the researchers considered factors such as the energy consumption of a computer running word processing software per hour. The comparison highlighted that AI systems like ChatGPT are remarkably more energy-efficient in generating text at a faster pace.
The study also evaluated per capita emissions in the United States and India to provide a global perspective on environmental impact. Results indicated that AI systems like Bloom and DALL-E2 were significantly less environmentally impactful compared to human counterparts in both countries.
While acknowledging the potential of AI to streamline tasks and reduce energy consumption, the authors emphasized the importance of considering factors beyond carbon emissions. They highlighted the need for a collaborative approach between AI and human efforts to optimize productivity while retaining control over the quality of output.
In conclusion, the authors underscored the significance of balancing AI integration with human creativity and control to navigate legal, economic, and societal implications effectively. The study aimed to contribute to the United Nations Sustainable Development Goals by promoting sustainable consumption patterns and addressing climate change challenges.
As the research continues to explore the intersection of AI technology and environmental sustainability, the authors remain optimistic about the potential of AI to enhance human productivity and creativity while minimizing environmental impact.
