In the recently concluded school year, a unique class of learners caught everyone’s attention with their hard work, improvement, and impressive articulation. Surprisingly, these learners are not human students but artificially intelligent chatbots. Despite their abilities to write poetry, summarize books, and provide answers with human-like fluency, these chatbots, such as Open AI’s ChatGPT, often struggle with math.
These AI systems can perform various tasks based on their acquired knowledge but encounter difficulties in delivering accurate math solutions. They excel in determining probabilities rather than executing rules-based calculations. The inherent flexibility of language poses a challenge when transitioning to the rigid structure of mathematics.
“The A.I. chatbots face math-related challenges due to their original design limitations,” explained Kristian Hammond, a leading computer science professor and AI researcher at Northwestern University.
Surprisingly, the world’s most brilliant computer scientists have unintentionally crafted AIs that lean more towards the liberal arts than math proficiency. This departure marks a significant contrast from the historical perception of computing, which has always been synonymous with advanced mathematical capabilities.
Since the inception of early computers in the 1940s, computing has been defined as “math on steroids,” with computers being unparalleled in their ability to perform tireless, rapid, and precise calculations. However, the latest AI creations prioritize linguistic finesse over numerical accuracy, reflecting a fundamental shift in the field.
Unlike traditional computing methods that rely on structured databases and rule-based programming, modern AIs prioritize probability assessment and language processing. This shift has transformed AI capabilities, making them more adaptable and versatile but also introducing unique challenges such as the math conundrum faced by chatbots.
