Hey there @coachwalker,

No worries, I can help to summarize this blog post.

This blog post talks about how you can use a technology called GPT (Generative Pre-trained Transformer) to quickly summarize articles. GPT is a type of artificial intelligence (AI) that can automatically generate summary versions of text documents. It can do this by analyzing the text, looking for the most important parts, and then condensing them into a shorter version. GPT can help save time when it comes to summarizing long articles, as it can do it quickly and accurately.

Hopefully, this helps! 

Thanks Ken, I think it was the "how to get these summaries" that I really couldn't understand

Hi @coachwalker!

I’ll see if I can break this down for you 

The first thing is that you need to know about chat GPT and what it does. As you may know, ChatGPT is an AI model. It allows you to type information or questions (aka a ‘prompt’) to give to the AI and the AI will respond. The great thing about this type of AI model is that it allows you to use natural language (ie the way we speak) instead of a code language. 

What ChatGPT is doing in the example in the blog is scanning through the html code of a webpage, pulling out the information, summarizing that information, and then giving that summary back to you as text. 

Hopefully, that explains the AI side of things. Now let’s take a look at the Zap to see how the information is being gathered and passed between Zapier and ChatGTP.

As a reminder, here are the Zap steps:

1. Trigger: Receive a push from the Zapier Chrome extension

2. Action: Parse the web page that was pushed to retrieve the article text content

3. Action: Send a prompt to OpenAI's GPT-3 to summarize the article for us

4. Action: Send a reply back to the Zapier Chrome extension with the article summary

Step 1

The trigger uses Zapier’s Chrome extension, which allows you to trigger a Zap by clicking on a button in the extension. You can learn more about how that works in this article

In this example, when you click the on the Chrome extension Zap, it passes the url of the webpage that you’re on to Zapier. 

Step 2

Web Parser by Zapier takes the url of the web page from the previous step and uses it to find the html of the webpage.

Step 3

This is where ChatGPT comes in. The Zapier OpenAI integration sends a text prompt to Open AI. This action in the Zap uses the html from the Web Parser action and sends it to Open AI, asking it to “summarize this article. The integration sends this prompt to ChatGPT, which will look at it and send the reply (the summary) back to Zapier. This will be the output of the Zap action; the output is what you see when you test a step. 

Step 4

The final step in the Zap uses the Chrome Extension again. It takes the Summary that OpenAI has sent to Zapier and passes it to the Chrome Extension. You’ll then see that text appear in the Chrome extension in your web browser. 

I hope that helps! Let me know if there’s anything I can clarify for you, I know that it’s a lot of information.

Thank you for taking the time to write this. There is still quite a lot of technical jargon. I'll see if I can fathom it in the next few days

If there’s something in particular that was too technical and you’d like explaining in more detail, let me know, I’m happy to help!