We Used ChatGPT to Write a Blog About ChatGPT: What that means for Instructional Designers

Posted by Liesl Christle & Grant Morin - March 28, 2023

header-picture

ChatGPT is a conversational AI model developed by OpenAI. It is a transformer-based language model that has been trained on a massive amount of text data to generate human-like responses to a wide range of questions and prompts. ChatGPT can be used in various applications, such as chatbots, content generation, and question-answering systems, to provide automated responses to user inputs. ChatGPT is capable of understanding context and generating coherent, relevant, and natural-sounding responses, making it an advanced and powerful tool for AI-powered conversational systems.

That’s how ChatGPT responds when you ask it to define itself. Pretty cool right, or maybe a little scary? OpenAI’s latest release of ChatGPT3 has built tremendous excitement, discussion, and in some cases nervousness, around the growing impact that AI chatbots could have. While AI isn't a new technology (have you talked with Siri or Alexa today?), the level of quality and performance that is being seen with ChatGPT is really what all the hubbub is about.

We wanted to do a little of our own exploration into its potential, especially when it comes to writing and instructional design. And what better way than asking ChatGPT itself about how it sees its own influence.

To start, accessing ChatGPT is easy (and free). You simply head to OpenAI (https://chat.openai.com/ ) and create an account with an email and password. From there, the world is your oyster in terms of the prompts and questions you can begin to input. One of my favorites that I've heard of so far is asking "How is instructional design like Encanto?" - it's a fun read if you have a few minutes. But there were some other things we were curious about….

 

Q: What are the limitations with ChatGPT and instructional design?

🤖:While AI language models like ChatGPT can be a valuable resource for instructional designers, there are also some limitations to consider. Here are a few limitations to be aware of when using AI language models like ChatGPT in instructional design:

Lack of domain-specific knowledge: AI language models like ChatGPT can generate responses based on patterns and relationships in large amounts of text data. However, they may not have the specific domain knowledge required to generate accurate responses to complex instructional design questions. Therefore, it is important for instructional designers to use their own expertise and judgment to evaluate the responses generated by AI language models.

Limited creativity: AI language models can generate responses based on existing patterns and relationships in text data, but they may not have the creativity and originality required to develop truly innovative and engaging learning experiences. Therefore, instructional designers should use AI language models as a tool to support their own creativity and innovation, rather than relying solely on them for instructional design.

Limited emotional intelligence: AI language models lack the emotional intelligence required to understand and respond to the emotional needs of learners. Therefore, instructional designers should use AI language models to support their work in creating engaging and effective learning experiences, but should also rely on their own emotional intelligence to create a learning environment that supports the emotional needs of their learners.

Ethical considerations: Finally, instructional designers must be mindful of the ethical considerations of using AI language models in instructional design, such as privacy concerns, bias, and fairness. AI language models may unintentionally perpetuate biases present in the data they were trained on, so it is important to evaluate the outputs of AI language models with a critical lens to ensure they are not perpetuating harm.

Overall, while AI language models like ChatGPT can be a valuable resource for instructional designers, it is important to recognize their limitations and use them as a tool to support, rather than replace, human expertise and judgment in instructional design.

 

We tend to agree with our AI-driven friend here. The role of an ID is still extremely relevant and necessary when looking to create impactful learning events. While ChatGPT can provide a basis for content development (we'll discuss this more below), it cannot yet determine what the needs of your learner are, how best to address those in the learning, and ensure active learning and critical thinking are occurring for them. The human-element of learning is still required to be done by - well - a human.

Getting to know your learners and wanting to design the best training that you can for them is the heart of a human-centered design practice, and how you develop empathy for your learners, understand their needs, and creating training experiences that support those needs.

Also - did you catch the part about “ethical considerations”? As ChatGPT, and other AI tools, grow in use and accessibility, so too does the concerns around accuracy, privacy, and biases. Currently, ChatGPT is more of a mirror – it reflects the data that we’re putting into it. So, while it is aggregating an immeasurable amount of data and then generating human-like responses, that data can also bring along fallacies and biases that were either part of the original data or that ChatGPT has created by attempting to teach itself. This means that like any other source material that you find on the internet, it shouldn't just be assumed to be 100% accurate. It's worth you doing the research to make sure that key points align with reliable information sources.

However, this doesn't mean that ChatGPT has no applications when it comes to writing and instructional design.

 

Q: How can ChatGPT enhance my role as an instructional designer?

🤖: As an instructional designer, you can leverage the benefits of AI language models like me in a number of ways to enhance your work. Here are a few examples:

Generating content ideas: One way to leverage AI language models is to use them to generate content ideas. For example, you can input a topic or concept into an AI language model like me and receive a list of potential subtopics, questions, or themes that can be used to create instructional content.

Personalizing learning experiences: AI language models can be used to personalize learning experiences for learners by analyzing their behavior, preferences, and performance data. This can help you design and deliver learning experiences that are tailored to the individual needs of each learner.

Automating repetitive tasks: AI language models can also be used to automate repetitive tasks such as grading, data analysis, or feedback. This can free up more time for you to focus on other aspects of the instructional design process.

Enhancing natural language processing: AI language models like me can also be used to improve the natural language processing capabilities of your e-learning platforms or instructional content. By leveraging AI-powered natural language processing, you can create more interactive and engaging learning experiences that can adapt to the needs of your learners.

Overall, by using AI language models like me, you can enhance your instructional design work and create more effective and engaging learning experiences for your learners.

 

Whew…that's a lot - but so are the ways that AI is creating opportunities for both creators of learning content and tools, as well as the learners. Hence why it is such a highly-discussed topic, being shared all over the internet.

But did you notice something with this response? It answered the question - sort of. The personalizing of learning experiences, or natural language processing, fall more into the category of technology developments and enhancements then it does writing and content creation. This showcases some of the limitations of ChatGPT, as the response is only as good as the prompt/question it is provided, as well as the data and content it is provided. As the technology that drives ChatGPT advances, these answers will probably get more and more accurate but it's still up to us to evaluate, edit, and enhance any output that we get.

One thing that is being discussed around ChatGPT for instructional designers is the use of AI to help generate content ideas - not replace content creation, but offer writers and designers a jumping off point when it comes to creating outlines, podcast scripts, etc.

 

So that’s what I tested out with my next prompt. I asked...

 

Q: Write me a compliance training on workplace behavior.

And you know what - it did! Below is a portion of the outline ChatGPT provided. It isn't a fully formed training with all the content pertinent to your organization, interactives, multimedia elements, etc. laid out -- but it is a start! Now you, the instructional designer, can step in and apply the human-element. Whether that's now taking the outline and brainstorming with SMEs, reviewing it with business partners to align with your organization's specific learning goals and objectives, or analyzing how best to deliver the content for your learners, ChatGPT has given you a place to start from.

 

🤖Workplace Behavior Compliance Training Outline:

I. Introduction

A. Explanation of the importance of workplace behavior and ethics

B. Overview of the purpose of this training

 

II. Definition of Unacceptable behavior

A. Harassments ( sexual, racial, age, etc.)

B. Discrimination

C. Retaliation

D. Violence 

E. Theft or Fraud

F. substance abuse 

 

III. Company Policies and Procedures 

A. Explanation of company's policies and procedures related to workplace behavior

B. Overview of the disciplinary process for violations 

C. Explanation of reporting process for incidents of the workplace misconduct

 

 

AI tools aren't a passing fad and as their technology improves, so will their reach and impact - Slack is just one of the latest to announce its own AI integration. It is true that these tools also bring continuously growing risks and concerns around privacy, plagiarism, and security. As with any tool, instructional designers and L&D teams should continue to monitor, research, and yes, even have fun and play with the tool to determine the best use case for themselves, their organization, and (most importantly) their learners.

 

Have you utilized ChatGPT yet? Share your thoughts and experiences below!

 

Topics: eLearning


Recent Posts