Skip to content

ameya-g-git/keep-it-real

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

46 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

banner

keep-it-real

A project created in 7 hours for WolfHacks 2024 hosted by Chinguacousy S.S.!

In a world overtaken by AI, where no piece of online media can be trusted, we need a shining beacon of hope in a wasteland of misinformation. We need something to Keep It Real.

Keep It Real is a Chrome extension that uses a combination of AI models and user-submitted feedback to determine the validity of information on the internet! By training our very own model on a dataset of fake news articles and headlines, our extension is able to provide users with a validity score for the page they're currently on, using a combination of prediction models and user votes. This system returns 95% accuracy in determining fake/AI-generated textual content in our tests! On articles that it deems as valid, Keep It Real will also provide readers with alternative perspectives and links to external news sources as to provide a well-rounded and balanced worldview on a given topic. With this extension, a more informed internet becomes more easily attainable.

Though the project has its fair share of bugs and incomplete features, it is an excellent proof of concept for the wide breadth of possibilities that come as a result of the intersection between user-generated content and AI tools. By effectively utilizing this combination of human knowldege and machine learning, both sides of this system can improve, allowing us to achieve a more educated future.

Frontend: HTML, TailwindCSS, JavaScript
Backend: Flask, CohereAPI, SciKit, NLTK

keepitreal_demo

About

a chrome extension created for wolfhacks 2024 that uses the cohere API alongside an AI model my team and i trained ourselves to detect fake news and ai written content.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • Jupyter Notebook 52.8%
  • CSS 29.6%
  • Python 8.5%
  • JavaScript 5.5%
  • HTML 3.6%