-- By IndraDan - 16 Dec 2023
One industry that language models have the potential to revolutionize is education. Khan Academy, one of the pioneers of effective online education, has already begun to introduce AI chatbots. With an emphasis on the ability to “personalize” a user’s experience, BigTech? intends to transform the educational landscape now. Yet, it is unlikely that the free-market will wait to identify the optimal approach for AI-based education. Instead, a trial and error approach with society is more likely. Thus, it is valuable to analyze the costs and benefits of these developments.
One distinction that may be relevant is certainty. Some academic subjects are structured around objective, definitive rules. In elementary mathematics, arithmetic operations with numbers rely upon these rules to come to a certain result. The guidelines of the subject matter are so clear that a student of these subjects needs only to become familiarized with the rules in order to “learn”. This example could be compared to a more nebulous subject like art. Art in general does not follow a formulaic approach - its value is often tied to the novel and creative processes involved in its creation. Attempts to establish rigid structure in these subjects is futile. Fundamental to the expression of art is humanity and its imprecise originality. For many, AI models, even when trained upon the most famous and storied pieces of art, are unable to capture the creative process and therefore unable to create products of value. Does the lack of clear principles in this process make AI unable to teach it? Unfortunately, very few subjects are developed upon infallible rules - everything exists on a spectrum of certainty. Even math begins to shift into the theoretical when pursued deeply enough.
Luckily, education is not a pursuit of certainty. Instead, it is an uncertain approach of making the student think, learning from experience (and mistakes) in a (somewhat?) guided fashion. From memorizing the capitals of the 50 states to developing the ability to speak in public, every form of education requires a student to engage with the content itself. Technology already serves as a valuable tool in this process, as it increases student’s access to resources. A successful teacher is one that can facilitate a student’s thinking process - by presenting content in a variety of ways and resolving questions that may appear. AI could be a convenient appliance in the larger education system’s arsenal - even if short-falls exist.
The student’s experience is the crux of this analysis. AI’s introduction to the education system should be heavily weighed against its impact upon the students themselves. While the aforementioned “pros” of an AI based education system are likely to drive adoption, there needs to be a corresponding review of what pains it may cause.
Individualization is a particular aspect to be wary of. Despite its benefits, inherently it would provide less opportunity for collaboration. Students confined in “personal” curriculums would no longer share their experiences with their peers - detracting from an important aspect of the current system. With Zoom school as a prime example, technology can isolate students. As a society, we will deal with the consequences of this isolation for years to come.
More questions will arise before answers are found. AI is likely to be a mixed-bag, coming with distinct advantages and disadvantages. The severity will depend upon how the technology is implemented. Caution and thoughtful analysis of these issues will hopefully help our educators avoid our prior failures.
Note: TWiki has strict formatting rules for preference declarations. Make sure you preserve the three spaces, asterisk, and extra space at the beginning of these lines. If you wish to give access to any other users simply add them to the comma separated ALLOWTOPICVIEW list.