How AI and Energy-as-a-Service Will Drive Decarbonization
This article is interesting as it seems to focus on an area where Artificial Intelligence oftentimes is not referenced, Climate Change. It seems a bit unintuitive, in contrast to other Big Data applications, where an AI is directly set to resolve some issue, here, the AI seems to serve a more tertiary role. Instead of directly 'solving' climate change, instead, the author suggests to use AI in order to make predictions about energy usage, which can then be optimized to be heavily efficient. This is especially useful when combined with some highly variable renewable sources of energy, such as Solar and Wind. By using AI to predict both the variable energy usage, as well as variable energy supply, we can create much more efficient energy system.
The interesting part about this article is how it uses AI as almost a secondary stance. Usually, AI seems to be thrown directly at a problem as it uses data to try and find a solution. For example, automated cars are almost entirely reliant on AI to do all the heavy lifting. Here, however, AI serves as a side benefit, almost an optimizer instead of the direct solution. I think this article serves to remind us all not to be limited about the scope of AI, and to continually think about how we can apply these in marginal areas to receive large benefits and value for society as a whole.
This article is really interesting because it isn't proposing any large-scale AI project that is overly complex, but rather points out that data that is already being collected could be harnessed as a tool to combat an important issue. The electrical grid management and usage of renewable resources is a difficult task to manage effectively, but by having technology take advantage of information we already have, it could be optimized to be more environmentally efficient. What is most interesting to me is how sometimes, just identifying how a software could be used and implemented is often more difficult than the actual creation process.