As learning and development professionals, we understand that training is not always the solution to every problem. The issue is that Chat TBT, our AI assistant, is not familiar with all the different models and frameworks we use as instructional designers and learning professionals. However, we can teach it. In this article, I will provide an example of how we can use a generic prompt to demonstrate the difference between using a model and our expertise in training.
Let’s start with a basic prompt: ‘Hi, I have a workplace performance issue I need help with. Please use the mega and pipe flowchart to assist me. Stay true to the model and ask me questions one at a time to find recommendations.’ The situation is that employees are not completing their time cards on time, and they may need training.
Before we proceed with ChatGPT using our models, it is essential to test its understanding. At the end of the prompt, I include the steps in the flowchart. This allows us to evaluate how well ChatGPT comprehends the mega and pipe flowchart.
Upon analyzing the output, we notice that ChatGPT follows the flowchart correctly for the most part. However, there are some areas that need correction. For example, it jumps right to training without considering other factors such as practice. To address this, we can provide additional prompts to clarify the model or start over.
In the revised prompt, instead of presenting ourselves as expert learning and development professionals, we include the mega and pipe flowchart in the prompt. This text-based approach allows ChatGPT to apply the model to our situation effectively. We begin with the performance problem, verify its existence, and then move on to expectations, resources, feedback, and consequences. ChatGPT asks follow-up questions to gather more information.
Next, we explore skills and knowledge. ChatGPT prompts us to assess if the employees can perform the task and if they have done it correctly before. It also asks for specific instances related to the performance problem. Based on this information, ChatGPT provides recommendations, including consequences, reinforcement, communication, and engaging in a conversation. It suggests less training, but as learning professionals, we may consider other approaches before resorting to training.
In conclusion, by using models and our expertise in learning and development, we can guide ChatGPT to provide accurate recommendations. It is crucial to ensure that ChatGPT understands the flowchart and can apply it effectively to address workplace performance issues.
