AI Assistants Under The Microscope: A Trainer’s Meta-Experience

I wanted to write a first introductory blog to explain what’s happening or happened this month. There’s been a lot of talk about AI and there are now a range of options on the market. I thought it would be interesting to join my voice to everybody else. Time to share some thoughts and options on AI Assistants.

First thing (but not the last time) to mention, I have to say I’ve really enjoyed this process!

I decided to be “meta” which is a different way or another way of saying ‘self-referential’ by using the relevant AI assistant to write each blog. As I said, I’ve really enjoyed it. It’s given me an interesting insight into working with them, what they are capable of, and of course prompt engineering.

Prompt engineering

If you’re not sure what this means, let me explain (a bit). It’s the main way that we interact with all of these AI assistants. It’s like an instant message, chatting with them via text.

When you need them to do something you need to give them that information. Ideally in as much detail as possible and that’s called a prompt. Anybody who’s had any experience with engineering, design, technology, or any manufacturing type thing, you know that the first version is not normally very good (a bit like pancakes). We see the flaws and ways that it needs to be improved. So what we do is try again or iterate, until we get to a final product that we’re happy with.

And that’s what we do with the prompts for AI.

You try out a slight tweak to what you did before. Then rinse and repeat. Until you end up either with something that’s exactly what you want, or very similar to it. With the very similar you can then do the final tweaks yourself.

And that’s exactly what I’ve done.

The Proof is in the Blog(s)

What you’re reading in each blog is the end result.

Note: This is only one that I have written without any AI assistance – it’s all me!

In keeping with the theme, I decided to take a copy of all of the prompts that I used, as I was creating each blog and share it with you. So you can see the process that I used in all it’s detail.

AI Assistants List

While there are a lot of options available, I’ve chosen to focus on the ones that I think are most well known.

Which ones have I chosen?

ChatGPT
Claude
Google Gemini
Microsoft Copilot
Jasper

The First Prompt

For all of the blogs, I started with the same first prompt:

“Hello, I would like help writing a blog post about using <AI Name>. It needs to be an easy read giving key information about pros and cons of using it and some suggestions to start playing around for small business owners. I’m an IT Trainer with c30 years experience and I like to write in simple terms, easy to understand with a little bit of comedy. Please write it in the first person with English UK spelling.”

The only thing I changed between each AI Assistant is the name, making sure that I used the right name on the relevant platform.

When you’re working across 5 different platforms that took a bit of concentration. 😄

Different Paths

Of course, after that first prompt, each AI went down it’s own path for the content.

Each post has it’s own linked files, which is only on the matching post and can be viewed or copied, free without the need to sign up for anything. You can see each additional prompt I used to change it, to tweak it and follow my process through. I hope that you find all of these interesting and will give whichever AI assistant or assistants a try.

Free AI Guide

Sigh, it’s been a busy period, because I created a new guide (free by signing up to my email list).

This free guide is your personal roadmap to mastering 4 top AI assistants – ChatGPT, Claude, Google Gemini, and Microsoft Copilot. Inside, you’ll discover 5 proven strategies to unleash the full power of these incredible tools and transform the way you work. Get ready to boost your productivity and enjoy personalized assistance tailored to your unique needs.

Mastering AI Assistants: 5 Must Know Tips for Effective Collaboration