Inspiration

Due to the ongoing situation in the Middle East, there were multiple rumors going around for example: Burj Khalifa caught on fire, Dubai Airport was destroyed, etc. Due to the misleading nature of these rumors we were inspired to create VeritasLocal.

What it does

The user can enter a common rumor circulating. The website can respond whether the rumor is Verified, Unverified or Misleading. A Verified Claim is one that is backed by multiple well known sources. An Unverified Claim is one that has limited information online. A Misleading Claim is one that is incorrect and has information online that proves that it is incorrect.

How we built it

While we were thinking of ways to build it, we thought of a solution. Using the help of various tools online like Gemini and Google AI Studio we proceeded to create a model which uses Gemini API Key and duckduckgo to search information online and checks the validity of the rumors.

Challenges we ran into

A few challenges we ran into were : Inaccuracy of model and the inability of it to assess most recent data. We managed to overcome this hurdle by feeding the model scraps of recent data and programming it to take recent data as priority.

Accomplishments that we're proud of

After a few days of diligent hard work, and continuously feeding the model data. We were finally able to create a model that can accurately decide whether a claim is Verified, Unverified or Misleading.

What we learned

We learnt how to create and use API keys. We learnt the importance of checking the validity of rumors. We learnt how to use AI tools like Gemini and Google AI Studio to help improve our project.

What's next for Veritas Local

With more planning and time, we want to expand our project to a 100% accuracy and make it more widely available and easy to integrate into a persons life to prevent the spread of misinformation in the world.

Built With

Share this project:

Updates