As AI tools grow in power, there are many negative and problematic uses in schools — but in my view, the new technology, if used correctly, can also be a helpful — and ethical — tool for improving student learning.
Among the ethical uses are math tutoring, generating practice problems and helping with research. These have all helped my learning.
As one of the most academically challenging subjects, math can be difficult to understand — especially when you’re taking high-level math classes as an underclassman. Often, the concepts are hard to grasp just from listening to lectures and reading the textbook; I have found some AI tools helpful for explaining concepts and solving problems.
AI’s visual analysis and mathematical capabilities are constantly improving. One of the leading models is Gemini’s 3.1-Pro-Preview, which is the most accurate at solving complex math problems. The model I found most helpful for everyday math aid is their newly released 3.1-flash-lite-preview model — it excels in speed and accuracy, since high school math doesn’t require particularly advanced AI models to solve.
AI can also be a great tool in a school setting for creating practice quizzes and mock exams. Gemini recently released NotebookLM, a helpful learning tool that lets users upload PDFs and textbooks. You can then ask questions about specific concepts, and it will explain them to you based on the understanding it derived from the files you provided.
Building on its advanced understanding capabilities, NotebookLM also lets users create practice exams and flashcards with a single click from the files you provide. It will then instantly create an interactive page of multiple-choice questions or questions that fit your desired practice styles.
Beyond math, AI can also be useful for research. The top AI models I have found most helpful are Perplexity and Gemini. Both excel at helping locate reliable sources, while also citing and providing links to the sources they used.
The pros of Perplexity lie in its ability to dive deep into sources, extract detailed information and compile it into a format understandable to users.
One downside is that Perplexity’s advanced research functions come with a monthly subscription fee of either $17, $34 or $271. As a high school student trying to make the most of what I have without emptying my parents’ wallets and getting a 2-hour lecture about saving money, Perplexity isn’t the go-to option for me.
But here’s where Google’s Gemini stands out once more. Due to its unlimited access to Google’s web and search results, as long as you prompt it to use a particular number of sources from reliable sites, such as those with “.gov” or “.edu” suffixes, its efficiency and flexibility are high.
And when it comes to Gemini’s pricing, it’s technically unlimited, as long as you have many Google accounts to rotate through whenever your daily prompt limit is up. Usually, I don’t find myself having this problem, as Gemini is pretty generous with its limit — and unless I am doing extensive research in one day, that limit is sufficient.
While some may disagree, I consider all of these uses of AI as highly beneficial to my learning and within traditional ethical boundaries.
To me, the line between ethical and unethical uses of AI is simple: It is the difference between using it to learn and cheat. The use of Gemini to help break down a complex calculus problem, and having it explain the process of solving the problem, allows me to learn the concept and apply what I learned to my homework. This applies to using it to generate practice problems, to act as a personalized tutor while it guides me to understand the material myself.
In terms of using AI for some research purposes, I think it’s OK to have it locate and provide reliable sources in the same way many databases do. This usage is ethical because I am using these sources to learn about whatever material I don’t understand. Using those links and researched material to submit as your own work for a works cited page would be a different story, but using them to understand material is totally ethical.
The line is crossed, rendering AI unethical, when it replaces a student’s own effort. This can be in terms of copying and pasting an AI-generated essay, or having it solve your math homework for you and then submitting its work while just copying its steps on your paper.
I draw the line of ethical and unethical uses of AI when it stops being a tool, but rather a shortcut to bypass learning.
Though ChatGPT used to be the most popular model, it doesn’t have the accuracy that Gemini has with its integrated Google-search-rooted responses now. I switched to Gemini around six months ago, and it has really improved and outperformed other AI models in terms of pricing and accuracy.
If you’re looking for extra help with your academics without paying for a tutor, I would recommend Gemini because it offers the most generous daily limits and is the most realistic for most high school students in terms of accuracy and speed.
