Quantifying Media Bias with Artificial Intelligence

I will create an AI program that quantifies bias in the media and then assign bias scores to raise public awareness around media integrity

Photo of Milo Knell
4 7

Written by

Additional categories (optional)

  • Environment

Date You Started Your Project Started


Project Stage: Select the description below that best applies to your approach.

  • Idea (hoping to get started in the future)

1. The Problem: What problem are you helping to solve?

Democrats and Republicans are politically polarized and less receptive to ideas from the other party. Climate change will be catastrophic if we are too divided to act, and people will continue dying from drug monopolies that are unchecked due to partisan bickering. Polarization is largely driven by biased news coverage: Americans read news with different presentations of facts and are 33% less likely to trust outlets favourable to the other party

2. Your Solution: How are you planning to solve this problem? Share your specific approach.

To help solve the problem of bias in the media, I am applying my skills and am developing an artificially intelligent computer program that reads news articles and rates the level of bias in the opinions and presentation of facts. Increasing public, and even media, awareness of this issue will publically pressure news organizations to change their standards of journalism, present less biased news, and ideally help to reduce the effect of media-driven political polarization. My AI will scan thousands of articles, identify biased presentation of facts, and calculate a polarity score for the article. My project is at the cutting edge of computational linguistics research. Identifying polarizing content is a very complex task, one that computer models alone remain incapable of achieving. To obtain accurate results, I will use the English language as a framework for the computer models I train, and use the connections between words to better understand how the meaning of individual words comes together to create a concept with a bias score. My project builds off existing research but has some key innovations (read more in the X-Factor section).

3. Personal Journey: What’s the story behind why you decided to start this project?

When my larger family gets together to celebrate, the conversation sometimes turns political, ruining an otherwise delicious meal. The members of my family are so set in their political views that productive arguments become prohibitive. Argument is a good thing, but not when animosity makes productive discussion impossible. Looking into why this is the case, I read that 90% of Republicans dislike Democrats and 84% of Democrats dislike Republicans; I also read how Walter Cronkite used to deliver the news to all Americans. This made me think about what changed and what could be done about our current situation. Given my desire to not sit idle and my ability to write code, developing an AI that can help identify biased news seemed like a good place to start. My work on this project has been very rewarding and I really look forward to continuing and expanding it in the future.

4. Selfie Elevator Pitch: Include 1-minute video that answers the following “I am stepping up to make change because...”

I am stepping up to make change because America is becoming more polarized, in large part due to biased news coverage. I have spent the last 5 months designing an artificial intelligence program to identify biased content in news. I will create rankings of the most biased organizations with two goals: to make people think about bias in where they get their news, and to pressure to be less biased. I hope to change America for the better through my project.

5. Example: Please walk us through a specific example of what happens when a person or group gets involved with your project.

If I partner with a person or organization that has platform, I would ask them to share my results and to help get the word out. I would also love people to help me in many areas of my project. It is small now, but I want to grow it bigger and bring other people on to work with me and help fill in the areas I am not as strong in. I am politically knowledgeable and skilled with technology, but I’m not as good with public relations or staff finding. As I have learned through my leadership roles in and out of school, it is more important to be able to delegate than to do it all yourself. Another capacity is simply getting the word out. Posting on social media, talking about it over the dinner table, or even thinking about it will all help.

6. The X Factor: What is different about your project compared to other programs or solutions already out there?

My project is at the cutting edge of computational linguistics research, and it has several innovations. I am training my AI and building its knowledge base (having it teach itself words) by having it read news articles to give it a political vocabulary. I am using pre-existing linguistics research to create rules about how the connections between words create meaning, but I am adding a few of my own specifically applicable for news articles. Finally, my program is optimized for speed so it can scan thousands of articles. I will publish a paper documenting how my AI works, and its results.

7. Impact: How has your project made a difference so far?

Once the data on bias is generated, I will publish a ranking of the most biased media outlets. Publicizing this ranking is intended to pressure media companies, and increase the public’s awareness of the situation and the effect it has on their political perceptions and outlook on life. If successful, the impacts of my project would be far-reaching: people will be more receptive to ideas from the other party, and vote for candidates with bipartisan aspirations. Argument is a good thing, but it becomes a problem when animosity between parties inhibits productive discussion.

8. What’s Next: What are your ideas for taking your project to the next level?

Should I win this competition, I will use the funding to rent powerful computers owned by Google to quickly analyze thousands of articles. Moreover, I would gain skills from the workshop; I would gain resources from T-Mobile and Ashoka; I would gain networking with others making a change and those who support them; and recognition to help me convince people to share my findings. If I am not selected, I will apply for grants to fund this project, and spend time finding people and organizations with platform who will share my findings. Regardless of the outcome, I will continue to vigorously pursue this project as I feel strongly that it can and will do some good.

9. Which of the following types of expertise would be most useful for you?

  • Public Relations

How did you hear about this challenge?

  • Recommended by others


Join the conversation:

Photo of Salma Kamni

Political polarization is arguably a top 10 issue in the U.S right now. Love the work you are doing with your project to eliminate it!

Photo of Milo Knell

Hi Salma,

Thank you so much! I really appreciate it, political polarization is a huge issue in the US and the world, and I can't wait to see where my project goes.

View all comments