About ten years ago, I had an idea.
I was watching friends and family share wildly different news articles on Facebook and Twitter. Some were outrage-driven. Others were clearly biased. More and more were outright misinformation.
It was frustrating. Not because people disagreed, but because the how of news was becoming harder to trust.
So I sketched out an idea. What if we could take a single story and synthesise coverage from multiple sources? We could use machine learning (now AI) to highlight the differences in tone, focus, and political leanings. Like a running sidebar, showing why and how each version was different.
My original concept was a unified, synthesised article, not just a mash-up, but something thoughtfully written. As you read, some lines would be annotated or highlighted. For example, “This line is neutral here, but Source A framed it more emotionally.” Also, “Right-leaning outlets emphasised this, while left-leaning ones didn’t.” A kind of live bias map layered over the text.
I never built it. I didn’t have the time or energy, or the know-how, honestly.
But a few years ago, I discovered https://ground.news, and there it was. Expressed differently, but built on the same principle: helping people see through the noise.
They break down coverage by political bias (from far left to far right). They show a “blindspot” rating to flag what you might be missing. You can follow specific topics. They even highlight factuality and media ownership. All these tools support more conscious, balanced media consumption.
I still think there’s room for inline commentary someday. Ground News is the best expression I’ve seen of the problem I was trying to solve. And I’m genuinely glad it exists.
Have you used Ground News or something similar? How do you keep your media diet balanced?


One reply on “An Idea I Had, and the App That Did It”
[…] looks real until you start poking at the walls.And here’s where media literacy tools really help. I’ve written about Ground News before, which lays out the same story across outlets with different political leanings. It doesn’t tell […]