Experimenting with ChatGPT and Virtual Training

AI promises to revolutionize information generation and consumption. While there are concerns about its uses and misuses, there is no doubt that, just as Google changed our ability to search for and find content on the Internet practically overnight, AI will have a profound impact on our ability to rapidly design, build, and participate in virtual training.

ChatGPT is an AI chatbot that was developed by OpenAI. Released in November 2022, it was trained on a large dataset (with information from 2021 and earlier) to provide human-like responses to prompts. The more specific the prompt, the better the response. But beware: ChatGPT “hallucinates” when it doesn’t have all the information.

It has been criticized for providing false information that appears to be factual. The user is wise to fact-check its output. And, because its dataset doesn’t extend beyond 2021, it can’t respond to questions about current events.

So, what can it do? Well, it uses a language-based model to allow for conversational interaction with the user. It works by predicting the next word in a given text, which is all based on the patterns it picked up from the massive database it used in its training process.

If you’ve not already experimented with ChatGPT, it’s time. To get started, open a web browser and go to chat.openai.com. You can create a free account there and start using it immediately. Here’s just one example of a prompt that I used:

Prompt:

Create an acronym to help remember key concepts about conflict resolution.

ChatGPT:

Sure, here’s an acronym for conflict resolution: PEACE

P – Pause and take a breath before responding

E – Empathize with the other person’s perspective

A – Address the issue calmly and respectfully

C – Compromise and find a solution that works for both parties

E – Evaluate the situation and reflect on what could be done differently in the future

It produced this solid acronym in seconds. If I was going to use it in a conflict resolution program, I would change the word “Compromise” to “Collaborate” to better reflect my conflict resolution model. I’ve spent a long-time creating acronyms when I need them, and the speed with which ChatGPT gave me something to work with was amazing. As a designer, presenter, and facilitator, the possibilities I see are endless.

In our next NetSpeed Nuggets session, we will explore the uses of ChatGPT in virtual training. From helping the presenter prepare, to supporting the producer troubleshooting a technical issue, to giving participants a creative tool for problem solving in breakout groups. Join us at this 30-minute, complimentary session, Training Hacks: Leveraging ChatGPT for Virtual Training, on Wednesday, May 10, at 1:00 pm ET / 10:00 am PT.

Registration is required.

Virtual Trainer Tips: Learning in the Flow of Work

More and more, learning professionals are tasked with providing the resources for people to learn in the flow of work. We are building learning ecosystems for people, giving them easy access to information and resources at the moment of need.

Recently I’ve personally experienced learning in the flow of work on a redesign project for a client. It involves repurposing what was a virtual-instructor led program to a self-paced asynchronous collaborative program. Our design team is creating 1 to 3-minute videos, Adobe Captivate elements, reading assignments, and knowledge check activities which will be uploaded to the client’s LMS.

It’s been a few years since I’ve used Adobe Captivate, but I decided to dive back into this authoring tool so I could help the team out. It’s been enlightening to say the least. But what I’ve noticed is that my favorite way to solve a problem when developing content is by watching YouTube videos, stopping them after a minute, and applying the recommended action to my own project. It doesn’t always work the way the expert tells me it will work. Then, I go find a video by a different expert to hear their perspective and recommendations. If all else fails, I may look for the answer in user documentation.

This gathering of information and examples, combined with immediate application and feedback is my favorite way of learning something. I see the results of my efforts immediately. Of course, at times, it can be frustrating when I’m banging my head on a door that isn’t opening. It’s helpful to have a technical expert on call who can stop at my desk, look over my shoulder and point out something I’ve missed. (That technical expert could be looking over my shoulder virtually too.)

As Talent Development professionals, we can think of our role as being curators of information, resources, and human interaction. Sometimes the learning assets are electronic, but often they are combined with human elements (a virtual coach, for example).