p/clarity-7
AI-powered news reader to help reduce media bias
Jeremy Yamaguchi
Clarity — AI-powered news reader to help reduce media bias
Featured
9
Clarity is an AI powered news reader to help reduce media bias and pop your filter bubble. We use GPT to summarize how the right, left, and center have been reporting on a given topic and highlight any differences between the narratives.
Replies
Jeremy Yamaguchi
Hey Product Hunt! Rohan and I are the founders of Clarity, an AI-powered news reader that's designed to combat media bias and pop your filter bubble. If you’re a sentient creature with an internet connection, you’ve probably come to recognize the growing tribal divide in our news media over recent decades. There are a few underlying drivers of this phenomenon: 1. First, the internet both commoditized the reporting of neutral facts and upended the newspaper advertising model. This forced media orgs to pivot to progressively more tribal fare in pursuit of clicks and subscribers. 2. Next, Congress repealed the fairness doctrine, which required the holders of broadcast licenses to present controversial issues in a fair and balanced way. This opened the door further to biased, one-sided reporting on critical issues. 3. Finally, social media came along with their personalized feeds, filter bubbles, and outrage-maximizing algorithms. These warped not only our own understanding of the world, but also changed the types of content media companies produce in response. We’re left swimming in our own private bias bubbles–with our connection to reality growing more tenuous by the day. Narrative has trumped fact, and we’re all the worse for it. What if there was a way to use AI to objectively analyze and present the news–one that used modern language models to aggregate and distill all the reporting on a given topic–helping you quickly see how the left, right, and center are painting the issues? That’s what @rohan_kshirsagar and I built over the last few months. We call it Clarity, and it’s an AI powered news reader to help reduce media bias and pop your filter bubble. Here’s how it works: • Rather than having to read through everything that’s been published to understand all sides of an issue, Clarity pulls together the top news stories and clusters them into topics (e.g., Conflict in Ukraine, Covid 19, Elon Acquires Twitter, etc). • We then use GPT to summarize how the right, left, and center have been reporting on a given topic, while highlighting any differences between the narratives. This saves you time, paints a more complete picture, and exposes any biases in coverage. • Clarity also scores the amount of attention outlets are giving a story, so you can see how individual topics are being exaggerated or ignored by the various tribal factions. It’s fascinating to see what each side chooses to emphasize, or where they are hiding the football. There’s a lot more to the product, including an aerial view of all topics, timelines for how topics trend over time, and more. We’re very early, and we have a ton of ideas for how we can improve this over time. We'd love it if you gave it a spin and left a comment with any thoughts, ideas, or feedback! ❤️
Greg B (Product Pair)
Amazing. My wife and I were just saying how we should build this, but now we don’t have to! Any plans for an app?
Jeremy Yamaguchi
@greg_productpair it’s definitely on the roadmap, but we’re mostly focused on adding more depth and utility to the current product to start. Got anything you’d want to see that would make this even more useful for you?
Miguel Acuña
Congratulations Clarity Team, a much needed tool, can't wait to see the same being applied to news in other places of the world, different languages, and cultures but a lot of polarization same as in the US.
Jeremy Yamaguchi
@miguel_acuna Agreed! We have plans to add more international coverage in the months ahead.
Ayesha Awan
Congratulations on the launch of your new product!
Rohan Kshirsagar
Around the time of Covid, the BLM Protests, the Convoy Truckers, and every other iron-hot story, I realized I was living in a media bubble. I couldn’t peel apart stories that were curated for my worldview from reality. Pew Research Center reported the extent of the US media polarization in 2020 and it’s only gotten worse since. I felt it myself — I was starting to trust my own sources less and less. I started studying my news diet by combing through the front page of multiple mainstream media outlets each day. What I found out was striking - The data showed routine and systematic bias of a category that most people didn’t talk about — topic coverage. Our news outlets bias us not just by how positively or negatively they talk about a topic, but by whether they talk about a topic at all. On top of this, outlets spin any topic too important to ignore. For the economy, for example, left-leaning outlets would spin the recession positively, and right-leaning outlets would do the exact opposite. --- Over the past year, after hundreds of conversations, thousands of data analyses, and millions of views on Reddit, Jeremy and I are sharing a new kind of tool: Clarity, an AI platform to reduce media bias, summarize current events, and make you smarter - useclarity.com Behind the scenes are some really exciting data-engineering, clustering, transformer embeddings, and GPT3.5, the large language model behind ChatGPT to give you a distilled view of the news, across the left and right, domestically and internationally.
kyungjin
Really cool AI tool. Congrats on the launch!
PactFact
I have also make a gaming app on this week boom.