Inspiration
For the past month, we have been diligently working on our argumentative essays at school. However, certain topics are more constrained in terms of available sources than others. Finding credible, unbiased information was difficult. Our plight led us to the idea for TrueScan.
What it does
TrueScan offers a wide variety of features, including the ability to detect AI-generated content and identify bias in writing. Additionally, it provides an overall trustworthiness score and explains the reasoning behind its assessment.
How we built it
We built this app by splitting up development between website and AI tools. After collecting data from a given website, we built text and image detection to see if the website was AI generated and used open AI to detect bias. Finally, we put it all together and embedded it in our scan feature.
Challenges we ran into
One challenge we ran into was APIs and referencing them. We avoided spending money as much as possible, which caused multiple free trials to run out. After debugging, we finally have it working.
Accomplishments that we're proud of
An accomplishment we are particularly proud of is our use of GitHub. All of our work was synced across GitHub and allowed us to collaborate very efficiently. This is the first time any of us have used GitHub and we are pretty happy with the result of our journey with commits, pushes, and pulls.
What we learned
Throughout the past 20 hours, we've learned many life lessons such as patience and collaboration. We learned to be patience, as we used AI to help train our model, which tended to be wrong but would periodically produce fruitful results. We learned collaboration skills, as the three of us were all working on one file, and if we didn't collaborate, we'd all have different variables and not know what the others were doing. However, we used our newly learned collaboration skills to communicate our ideas and our work, to make sure we were all on the same page.
What's next for TrueScan
What’s next for TrueScan? We will work on thoroughly finishing our website, and work on finishing chrome extension to make the process even faster. Finally, we will annotate the actual html to show what specific pieces of info triggered our scan.
Built With
- beautiful-soup
- flask
- html
- javascript
- json
- openai
- png
- py
- python
- requests
- txt

Log in or sign up for Devpost to join the conversation.