Thursday, August 21, 2025

Answer: Best ways to do just-in-time learning. Part 2: conversation with a chatbot

  As mentioned... 

A woman studying a complex anatomy topic. P/C Pexels.com by RF._.studio _

      

There are a huge number of methods to quickly learn something, especially today.  This is the next in a series of posts about methods I've tried.    

Remember the Challenge: 

1. What are the best AI-inspired (or AI-driven) new Micro- and Meso-learning that you've found?  Have you actually tried this method?  How well has it worked out for you? 

Last week I wrote about just searching for pre-existing tutorial content.  That often works quite well and is a good first place to start.  

But sometimes you want something that's bespoke and tailored to you interests at the moment.  Here's a way to use the chatbots to do that.  

You can put both ChatGPT and Gemini (and many others) into chat mode that lets you have a conversation with the AI. Here's my example using Gemini.  To start the conversation, click on the button with the vertical lines (lower right) to get started.  



The key idea here is to ask the chatbot to read you an essay on the topic of interest.  Here, I continue the quest to learn more about kelp with my prompt: 

[Give me a 500 word essay on the natural history of kelp along the California coast]

Note that it reads around 120 words/minute, so I'm asking for a 4-minute reading. Ask for 1000 words if you want a more detailed version. 

Be sure to be as specific as you need.  Note that I wanted info about the "natural history" and specifically to the "coast of California."  (I know that kelp lives in many oceans, but I have a parochial interest in local kelp. See picture at the end of this post.)  

NOTE: As you listen to your topical essays being read aloud, note that you can interrupt it and ask a follow-up question.  (Such as, "what is the role of pollutants in damaging the kelp coverage?")  You don't need a special hot word (like "Hey Google!"), just start speaking when you want to drill into the topic. 

After you've done that,  you can say something like "... let's go back to the previous topic."  It really does feel like a conversation with a subject-matter expert. 

Of course, each of the chatbots will create a somewhat different essay. Here's the one that Gemini created and the essay ChatGPT created.  

Here's a visual for you (typography by me):  

Click to see in full-size
 
Notice that Gemini wrote 407 words, while ChatGPT wrote 530 words.  Gemini neglected to mention bull kelp at all, while ChatGPT gives it a prominent place.  ChatGPT also included information about Seasonal Cycles and the historical Human Use of kelp in California as well.  For my money, a much better job. But some might prefer the lighter touch of Gemini.    

I've found myself asking for short essays on topics that I'd like to know more about, especially when I'm about to do something (visit a museum, attend a seminar, go to a particular part of the city).  

As remmij pointed out, you can even use this chatbot conversational style to ask questions about learning method.  When I asked the bot: 

[tell me about the Feynman technique for effective learning] 

Gemini not only told me what it was, but then jumped into "Feynman teaching mode" on the topic of California kelp.  

But we'll talk more about that next time.  

 

SearchResearch Lessons


This is Part 2 of our "how to learn rapidly" series.  We learned that: 

1. Chatbots can read on-topic essays to you. Use this as a way to delimit what you'd like to learn.  

2. Think of a chatbot as a way to have a conversation with the material. In particular, you can drill down deep into a topic in a very natural, conversational way.  

But AS ALWAYS validate what it tells you.  Thus far, the conversations I've had all seem pretty high quality and accurate.  But that's not a guarantee.  If you have questions or hear something that sounds a bit...pardon me... fishy... be sure to check!  


Diving in the giant kelp in the Channel Island, California. P/C Dan.


Keep searching!  

No comments:

Post a Comment