The Voice of AI

Author(s): Sarah Cordivano

Originally published on Towards AI.

And how it creates overconfidence in its output

Non-members of medium can read this story for free through this friend link.

In the last year, ChatGPT and similar tools have written a fair amount of the content I see on Medium and LinkedIn. I’ve noticed these AI tools have a very distinct and recognizable voice. This is why they are so easy to spot in the wild. This is especially true on LinkedIn, where it feels like there’s an ongoing race to the bottom for low-quality content in search of claps and hearts. But what is the voice of AI, and why does it sound the way it does?

Photo by Cash Macanaya on Unsplash.

ChatGPT and similar tools produce text with a formal business tone, full of clutter and phrases that people never actually say aloud in real life. Seriously. Ask yourself, when did you last use the word “crucial” in real life? And yet, ChatGPT loves that word, along with many more, like: imperative, paramount, vital, definitive, moreover, furthermore, aptly, thusly, therefore, adequately, shall, embrace, engage, leverage, on and on. Those are words that I never hear in actual conversation, not even on scripted television! Yet they are everywhere in ChatGPT’s texts.

It’s almost like ChatGPT is a parody of… 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