Don't do this when using Generative AI models and tools

Most Generative AI tools available to the public (ChatGPT, Claude, Copilot, Perplexity, Gemini and the many platforms that use their models) are designed to optimize for speed. For the sake of providing a quick response to your prompts or questions, they generally have to sacrifice creativity, quality or reliability (because those take time) to give you a faster response.

To be speedy, but sound “smart”, the models and tools have to fill a potentially short response with a bunch of nonsense crap and fillers to make it sound more… intelligent.

For example, some of the most common words (fillers) used by these tools are: nuanced, multifaceted, weave, fabric, underscore, pivotal (damn it, I hate the word pivotal now), among the many others. These words add zero value, but they make a response look longer and more thoughtfully crafted (by the way, thoughtfully is another filler word, but this time it was written by me… a human).

In addition to that, at least in ChatGPT (I have not observed the same pattern in other models and tools), some sentences always start in the same way: “In the rapidly changing landscape of modern business…” or “In today’s rapidly evolving business environment…” (by the way, this last sentence is the first sentence of one of the articles of the “gurus” of HR Technology… go figure).

Frankly, I don’t think a lot of humans would ever start a sentence like those above. But if they do, I am certain that they wouldn’t start ALL their sentences that way.

Don’t ever, ever, ever leave a Generative AI generated response go unedited into your writing.

Those of us who know about AI will know that you didn’t write your content or that you didn’t at least take the time to edit it (like that “guru” who didn’t even edit the first sentence of the first paragraph of his article). But, even worse, search engines like Google will pick it up and penalize it because of its low quality (and the low quality is not because the content was generated WITH AI… but because it was generated BY AI --- see the difference? With or By… not the same – focus on creating high quality content WITH Generative AI, not BY Generative AI).

You must edit it. If not for the very sake of making it sound human, which in itself is critical and valuable, at least to remove the nonsensical crap that adds no value.

Instead of prompting ChatGPT to write something for you without any context or format, do this instead:

1. Ask to avoid filler words

In addition to what I explained in my first post of this new "AI in HR" newsletter (act as an expert, confirm you understand my request, ask me questions), I clearly tell ChatGPT or any other tool:

Do not use any of these words in your response: “nuanced, multifaceted, weave, fabric, pivotal” and do not start with “in the rapidly changing landscape of modern business” or anything like that.

2. Provide your writing style

Provide some examples of your writing style and ask your preferred Generative AI tool to confirm it understands the style, format and voice. I say: “I will now share with you some links and paragraphs that show you my writing style. Please check them out and confirm you understand my style.”

3. Edit, edit, edit -> to improve

Again, don’t do this: leave a Generative AI generated response go unedited into your writing. Create high quality content by collaborating WITH your preferred Generative AI tool. It is you WITH it.

By the way, I asked ChatGPT (Model 4) and Claude (Opus – premium level) what they thought (“thought”) about my write up above, these were their responses:

By Enrique Rubio (he/him)

(Note: I write fast, speak fast and read fast. Sometimes - often times - I don't see the typos! Please let me know if there are typos or grammatical misconstructions in my write ups. It happened to Shakespeare, it certainly can happen to me. Do so with kindness, compassion and grace, though. I appreciate it!)

ENROLL TODAY: Mastering ChatGPT (For HR Professionals)

Introducing the first, ever, LIVE Certificate Program focused on mastering ChatGPT for strategic and tactical HR.

The "Mastering ChatGPT (For HR Professionals)" mini Certificate Program (6-weeks) is strategically designed to equip HR generalists, specialists, and leaders with the knowledge and skills needed to integrate advanced AI technologies, like ChatGPT, into their HR practices effectively.

Spanning the essentials of AI in HR, tactical and strategic applications of ChatGPT, and innovative HR practices, this curriculum ensures participants are well-equipped to enhance recruitment, onboarding, culture building, and more through AI.

Enroll today (enrollment closing on Friday, April 26).

Download Our Full Guide

Click here to download.

?