Hey there!

For those who are new to the newsletter, I talk about the integration and application of AI in the modern world.

I’m calling this edition ✍️ Do Your Homework. We’re going to discuss how AI can improve the lives of teachers and students. Needless to say, I’ll be doing another edition on this later down the line.

Here’s what’s in store:

• A Facelift: As you can see, I put in some time in building the aesthetics to this newsletter. Let me know what you think of it!

• The Writing: Both ends of the classroom roles are difficult in their own aspects. I investigate a few different ways in which AI can help both students and teachers with achieving their objectives.

• AI Tools: As always, I’ve included 3 tools I think are really cool and can be of use for students, instructors, and even entrepreneurs!

• Midjourney Images: I’ve got 3 in store for you for this edition. Prompts included!

It’s my goal to hit 100 subscribers by the end of this month (July 2023). You can help me do so by referring your friends - if you’re getting this in your inbox, scroll to the bottom of this email. For those on a webpage, send them the URL to the webpage.

Without further delays, grab your cuppa joe and enjoy…

 New here? Grab a cup of coffee - we’re talking AI here. The content that’s in this newsletter is curated towards the integration and applications of AI in our lives, so if this sounds like your kind of thing, hit that “subscribe” button below!

I remember very explicitly my 7th grade math teacher said “you will never, in your entire lives, carry around a calculator in your pocket, so you will need to learn how to do math on pen and paper”.

Sorry to say it Mr. Anderson, but we’ve got a calculator in our pockets in 2023.

Calculators changed how we learn mathematics this day and age. It took a long time for these devices to make their way into classroom due to accessibility reasons, but they've now reached a point where they're a crucial tool on the student’s tool belt.

I think the same will be true for AI. Generative AI (i.e., GPT-backed applications) will become a tool on not only the student’s tool belt, but also on the teacher’s. Ethan Mollick (@emollick), a professor at The Wharton School, has written an article that examines the different kinds of assignments in the classroom and explains how AI plays a role in the evolution of these kinds of assignments.

I’ll leave the detailed explanation to the article, but I want to branch from it and explore how AI can be (and currently is) implemented to support not only students, but instructors as well on different aspects of the education environment.

Let’s suppose we have a student who is trying to solve an algebra problem. A problem-solving workflow paired with an online math agent to grade their work may look something like this:

In this sample workflow, we see that the online math agent doesn’t provide feedback (yes, such systems still exist!). This leads to frustration and students becoming demotivated to learn the topic since they do not understand where and why they went wrong; this is key.

With generative AI taking off recently, an AI math assistant backed by GPT will provide more insight as to where they went wrong and helps the student understand why they made an error, leading to a less-frustrated state on the student’s end.

In the same scenario, if the student were to have an assistant backed by GPT with a “Get AI Assistance” button on screen:

I should add in, there is room for abusing the “Get AI Assistance” button by spamming it every step, but this is under the assumption there is no safeguard against it at the application level. Not to go down a rabbit hole, but I think this opens the door to explore the topic of “how can we get students to be less reliant on AI assistance at this level long-term?”.

This is just one example where AI can help students. There’s a few others:

• Problem set generator: Students learn best when they receive personalized and tailored explanations. In a classroom setting, problem sets are often distributed to practice specific topics, but this assumes that all students are at the same level. In a case where a student falls behind, an AI-powered problem set generator can play a crucial role in helping the student catch up by generating them problems to revisit earlier topics.

• Organization: Students have to juggle multiple classes during the semester. Each instructor has their own unique way of delivering the content of the course, grading, and instructional method. In addition, the class can be held either in person, hybrid, or online synchronous/asynchronous. This decoupling across classes and instructors can overwhelm students, especially during finals. As such, tools such as Notion AI* can be created to help students remain organized.

Now, let’s shift our focus and discuss how instructors can use (and are using) AI. Here’s a few examples:

1. Polishing lecture content: In a scenario where instructors aren’t confident on a particular lecture, the instructor can give the lecture verbally, as if they were talking to their class, to a speech-to-text AI tool. From there, they can feed the transcribed lecture to a large language model (LLM) to get suggestions for improvement. This will result in increased confidence with the lecture material and also establish a feedback loop for instructors to provide higher-quality content.

2. Automation: From my experience with college, my professors spent a lot of time grading the student’s material and creating the coursework. By having AI automate most (not all!) of the grading process and expedite the coursework content creation, they can free up that time to have more office hours, for instance, for their students.

3. Student inclusion: It’s quite common that classes at universities have students whose first language isn’t English. A tool such as the Microsoft Presentation Translator can help students understand live lectures by translating the lecture in real time. This leads to more of an inclusive academic environment within the classroom setting.

I did discuss some of the limitations of generative AI (specifically, LLM’s) in Edition 1. As such, any integration of LLM’s should take into account the limitations mentioned in the edition.

That said, I do wonder what Mr. Anderson is saying about AI…

Are you an instructor who uses AI in their course? Are you a student who uses AI for their coursework? I’d love to hear from you! Simply respond to this email and it’ll appear in my inbox (psst, you may be mentioned in a future newsletter 👀)

Any links with an asterisk (*) are affiliate links

1. Courseau

Easily create an online course from a book, podcast, notes, or even a video into a course in a matter of minutes.

I added my Introduction to HTML course to see how AI handles a pre-built course. It was able to restructure it and display it in an easily digestible manner. I recommend checking out this neat tool on their website. The developers are actively developing it and would love to hear feedback! (Share on Twitter)

2. Albus

Explore any topic using GPT. Build a board and share it with others (here’s mine!).

This tool can be useful in and outside of the classroom for learning and teaching. For instance, one student will build the board on a subject they’re knowledgeable about, then hand it off to another student, who will then present the board. Check this tool out on their website. (Share on Twitter)

3. MagicSchoolAI

Use the power of AI to quickly write lesson plans, streamline communication with parents, and create quiz questions.

This tool is designed to support teachers with content generation, saving them hours of manual work. Check this useful tool out on their website. (Share on Twitter)

The University of al-Qarawiyyin, home to the oldest working library in the world, is located in Fez, Morocco. This library was founded in 1359 CE by a woman named Fatima al-Fihri. It underwent renovations in 2016 to prevent further decay and has since opened its doors to the public.

Did you know…?