I Built an AI Tool to Expose the Lie My Blog Was Telling

Author(s): Mukundan Sankar

Originally published on Towards AI.

Image created by the author using ChatGPT

I was staring at my computer screen and felt confused. I had written fifteen hundred words, and I had looked over everything to make sure it was right. I chose each word carefully to make my writing clear. It should have made me feel good, like I was ready to share something really nice. But instead, it didn’t feel right, like my smile wasn’t reaching my eyes, i.e., the feeling when we don’t feel real joy.

I started to wonder: Was I being too formal? Too distant? Was my blog talking at readers instead of actually connecting with them? That question kept gnawing at me, making it hard to shake the doubt lingering in the back of my mind.

So I did what any obsessive writer might do.

I ran my own blog post through two tools — one trained to detect emotions, the other built to read like a human.

And what came back?

It made me rewrite the entire thing.

I created a tone analysis app built for writers like me.

I upload my blog. The app breaks it down paragraph by paragraph. Then it passes each one through two different tone detectors:

One powered by emotion classification.One that describes… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI