• Moodboard
  • Posts
  • A checklist for AI content that sounds more human

A checklist for AI content that sounds more human

Copy/paste this at the end of any prompt to stop sounding like ChatGPT

QUICK NOTE BEFORE TODAY’S EDITION: 

Thank you to everyone who has already registered for our first cohort of The AI Content OS!

We’re only 11 days away from the kick-off, and I’m genuinely stoked for you to experience the curriculum we’ve been putting together.

For this first cohort, I’ve made the decision to place a strict limit on the number of seats. Attendees can expect lot of one-on-one support from myself and the Moodboard team as we build out your personalized AI-assisted marketing engine.

We only have 6 spots left before we close registration, so if you want to join this first cohort (at the lowest price we’ll ever offer for the program), register now and claim your seat.

Hello everybody, welcome back to Moodboard 🌴

Writing to you today from Rome as my wife and I finish up our "babymoon" in Italy.

The first addition to our family (a baby boy) is arriving in January, so we're getting our last trip in together before things get too crazy.

Today, in another short & sweet edition, I wanted to share a checklist that I've been adding onto the end of any AI-content prompts to ward off the dreaded "AI-isms" that immediately give away a piece of content as authored by ChatGPT or Claude.

Let’s get to it 👇

The Checklist

Everyone knows the bad rep the em-dash has been getting lately, but there are some less obvious (but equally prevalent) phrases, rhetorical habits, and narrative tendencies that LLMs tend to rely on when crafting content.

As our team has familiarized ourselves more and more with these little annoyances, we've created a "checklist" of sorts that we can simply tag onto the end of any content generation prompts to mitigate how many of them show up in our first drafts.

Here's our running list for you to copy/paste:

AI-ism Elimination Add-on

Write in a clear, direct, and compelling style. Avoid filler or gimmicky phrasing. Specifically:

Do not use rhetorical flourishes like:

  • "It's not just [X], it's [Y]."

  • "The best part? …"

  • "The secret? …"

  • "Sounds impossible? It's not."

  • "What if I told you…"

  • "Here's the thing…"

  • "Let's be honest…"

  • "The truth is…"

  • "At the end of the day…"

Do not use faux-dramatic "staccato" phrasing like:

  • "No fluff. No filler. Just results."

  • "Big ideas. Small words. Maximum impact."

  • "Simple. Clear. Effective."

  • "Stop guessing. Start winning."

Do not start with clichés like:

  • "In the ever-evolving world of…"

  • "In today's fast-paced…"

  • "Gone are the days when…"

  • “This underscores…”

Additional restrictions:

  • Do not use corporate-y buzzwords or jargon.

  • Do not rely on bullet points or lists unless explicitly asked.

  • Do not pad sentences with unnecessary adverbs.

  • Do not overuse conjunctions like "Furthermore," or "However."

  • Do not use boldface, emojis, or decorative formatting.

  • Do not add summaries like "In conclusion" or "In summary."

  • Do not use em-dashes for dramatic effect.

  • Keep sentences natural and flowing, not choppy or over-engineered.

Instead:

  • Be concise and straightforward.

  • Write with natural cadence, like a skilled human writer.

  • Favor clarity and rhythm over cleverness.

  • Aim for clean, confident sentences that sound human and not over-polished.

The next time you're using Claude or ChatGPT for any content creation, just tag this checklist onto the end of your prompt.

It should drastically reduce the number of these tics that show up in your drafts 👍️ 

Try It Out

What do you think? Am I missing any?

If there are any AI-isms that you've noticed that aren't included in the above list, respond to this email and let me know. I'll add them to my own list (and feature them as a note in next week's edition).

Try it out for yourself, and let me know the results!

Until next week,

Ryan