During the summer of 2023 (article written August 2025), I contributed to research at Loyola regarding how ChatGPT understood CS2 data structures (defined from Pearson, the publisher, as the second Computer Science course in common curriculum–CS1 is akin to AP Computer Science A).
I was given the task of evaluating responses from ChatGPT and compiling them on https://curricula.cs.luc.edu/15-cs2/content.html. I contributed to the sections on Maze search algorithms. I found it fun to take my theoretical knowledge from my Data Structures class and pair it with ChatGPT to further enhance my learning.
Bi-weekly (obviously meaning once every fortnight) meetings helped guide the project and what direction to take when researching.
Credit to seanperfecto on GitHub for the above gifs.
I’ve compiled my chats with v3.0 into convenient markdown format:
- Iterative Maze Solver Example
- BFS, DFS, and A* algorithms among others
- Recursive Maze Solver Example
- Memoization, multi-threaded approaches
- Meta Researching ChatGPT
- A meta-level analysis where I ask the AI how it would research itself
- Family Ancestry Tree in Java
- Asking v3.0 how it would design an ancestry tree to find relatives
I believe they were listed on the Google Search index at one time in July 2025 due to them being shared with my reserach cohort as a public link, but OpenAI has since fixed that issue. Regardless, nothing sensitive was ever shared.
You can see how v3.0 was hallucinating (or just incompetant) back in 2023:
User:
Is the maze in your example able to be solved?
ChatGPT:
Apologies for the confusion, but upon reviewing the example maze, I noticed that it is not solvable.
The maze has an isolated wall in the middle, blocking any path to the exit.
Here's the maze for reference:
* * * * * * *
* *
* * * * *
* * *
* * * * * *
* *
* * * * * * *
This maze has no exits nor entrances, and its funny for the AI to point out the wall in the middle as it would not obstruct any path in the middle.