| Proj No. | B2155-251 |
| Title | Modelling socio-economic impacts of the rise of AI |
| Summary | AI is currently on the rise in its usage for various tasks, from consumer aligned word prompts, and chat bots, to business aligned pursuits of inventory and machinery maintenance and efficiency tracking. AI uses cloud computing and hence data centers, to allow companies to attempt to automate processes, without necessarily investing in hardware and software in-house. Data centers have been known to consume an immense amount of power not just for computation, but for cooling and other power overheads, causing recent estimates of power usage in 2023, in the united states, to be about 4.4% of its total electricity usage (~176 Terawatt hr ).Estimates on what that could look like in the future have varied between 500 TWh to 860 TWh. Suggestions have been made to improve the sustainability of this exponentially increasing use of AI, through the set-up of data centers near green-energy production sites to shift their source of energy to renewable sources, as well as shifting to the use of liquid cooling more widely. The current amount of water cooling being done resulted in the use of about 550,000 gallons of water daily to cool down Google's hyperscale data centers in 2023. Due to these incredibly high amounts of energy and resources needed to manage and maintain these facilities, often data centers are housed together, close by and set up at bigger scales, impacting the areas they inhabit in incredibly localized ways. This project chooses to explore how the data centers being set up could affect the areas they inhabit and how the localized effects can impact local lives, as AI projected to experience a compound annual growth rate of about 40%, and experience a shift towards hyperscale wholesale operators like Google and Microsoft, (70% in the near future) . These impacts would be modeled through utility expenses, and overall life quality by extension, of the various social classes in an area, along the way this impacts the number and spread of how data centers are set up. |
| Supervisor | Ast/P Matthew Foreman (Loc:S1 > S1 B1C > S1 B1C 77, Ext: ?) |
| Co-Supervisor | Dr Le Chencheng (Loc:N2, Ext: ?) |
| RI Co-Supervisor | - |
| Lab | Photonics I (Loc: S1-B3a-08) |
| Single/Group: | Single |
| Area: | Intelligent Systems and Control Engineering |
| ISP/RI/SMP/SCP?: | SCP: Le Chencheng Lecturer Asian School of the Environment ccle@ntu.edu.sg |