Скачать или смотреть Writing with AI Workflows - Perplexity, Microsoft Copilot, ChatGPT, Grammarly

When Twitch got popular, I was confused.
Would anyone really want to watch someone else play a game?

But the esports market is expected to be over $4 billion this year,
And Amazon acquired Twitch for almost $1 billion in 2014.

So, I thought I'd share my prompting and AI workflow in a video to explain my processes and see if there's a market for watching people use AI. (Just kidding, sort of 😆).

Anyhow, here's my AI creation process for a blog or an article:

First, I like to do some research on a topic.
For this example, I am researching the latest trends in Retrieval Augmented Generation (RAG).

I use both Perplexity and Microsoft Co-Pilot here.
There are no fancy prompts, just, "What are the latest trends in Retrieval Augmented Generation?"
Once I look at t outputs I know what I am looking for.
Sometimes I may create an outline but for this example I just copy and paste from Co-pilot.

Then I use a saved prompt.
This one is one of my Superprompts which is interactive and walks the user through the process of creating the blog and social posts by conducting and interview.

I am using a Chrome extension, PromptForge to save my prompts so I don't have to type anything, you'll see it on the righthand side of my screen in the video.

Then I cut and paste the results from my Microsoft Co-Pilot search in the interview when ChatGPT asks what I wanted the topic of the blog to be.

Once I finish the interview ChatGPT generates the blog and social posts.
Then I copy the past into Wordpress for the blog, and edit it with Grammarly as a proofreader.

And I make sure that the article is quality, not just AI-generated blather, though it may not pass an AI generation detector, if the article isn't good, I don't hit publish.

Then I cut and paste the LinkedIn and X thread into Taplio and Tweet Hunter or Hypefury to schedule the posts and once again I edit them with the help of Grammarly.