Inspiration
We wanted to combat concerns about reading comprehension -- specifically logical fallacies -- by using AI-generated examples embedded in a video game with the intention of teaching students in an engaging and dynamic way.
What it does
Our program is a working prototype for game that generates examples of logical fallacies in silly contexts (i.e., the moon is made of cheese) and asks the user, presumably a student, to pick the logical fallacy that is used in the situation.
How we built it
We used Python and the openai library.
Challenges we ran into
We tried three different APIs. ChatGPT-3.5 had issues with payment and Phi 3 mini 4k Instruct refused to provide examples of logical fallacies because of ethical reasons (it has safety checks that prevent it from lying, even in harmless situations; this is a good thing!). We finally settled on Facebookâs Meta Llamma Instruct.
Accomplishments that we're proud of
We are proud of our creative concept. We believe that our project has the potential to help reading comprehension and critical thinking in a fun, engaging way.
What we learned
AI has its challenges, but it is surprisingly simple to implement, especially in Python!
What's next for The Moon is Made of Cheese
- Expand UI/UX with website
- Add a high-score tracking capability.
- Expand to concepts outside of logical fallacies with different game modes (example. theme analysis, grammar errors, etc.)
Built With
- openai
- python
Log in or sign up for Devpost to join the conversation.