Edovate Learning Corp. is forward-thinking and embraces using AI tools for various purposes. Prime candidates for leveraging AI tools are creating efficiencies in the documentation of standards alignment (including maintenance of that alignment) and tools like ChatGPT to reduce time doing research and authoring.

More information can be found in Generating Instructional Text.

Using these tools comes with requirements, expectations, and cautions. While Edovate will monitor this changing technology and may update its policies, ChatGPT 3 and ChatGPT 4 are detailed here.

ChatGPT

ChatGPT use is acceptable and will return writing on various topics; however, that writing should never be considered the entity of submitted content. AI can generate anywhere from 50-80 percent of lesson text, but Elephango contractors must refine and improve work for a complete submission.

Cautions and Expectations

Edovate contracts stipulate that submitted work must be 100 percent unique. While ChatGPT’s terms of service allow for for-profit use of chat output, it does not guarantee that output is unique. Users submitting the same prompt could receive the same response – or a response that’s virtually the same.

Fact-checking and revision is required. ChatGPT fully acknowledges the results provided in a chat may contain errors, bias, and other writing that would be undesirable or inappropriate for Elephango students.

Do not submit Edovate’s intellectual property or personal or sensitive content to ChatGPT. Everything that goes into the system remains in the system.

You must include the ChatGPT URL for any chats utilized in creating content near the pertinent content. (Example at the end of this article.) You may need to leave the chat window and return to access a unique URL.

At the close of your chat in ChatGPT, ask it to cite sources for the information. Check those sources for appropriateness and accuracy before submitting the work order.

ChatGPT Examples

No modifications were made to the content or style in the following chat with ChatGPT.

Chat Entry

“What are some best practices for using ChatGPT?”

Chat Reply

“As an AI language model, ChatGPT can be a powerful tool for generating text-based responses to a wide range of questions and topics. However, there are several best practices to keep in mind when using ChatGPT to ensure the best possible outcomes:

1. Be specific in your questions: ChatGPT works best when given specific questions or prompts. Avoid vague or open-ended questions that could lead to irrelevant or confusing responses.
2. Provide context: When asking a question or giving a prompt, provide as much relevant context as possible. This can help ChatGPT better understand the intent of your request and generate more accurate and relevant responses.
3. Verify information: While ChatGPT is designed to provide accurate information based on its training data, it is still important to verify any information provided by the model before relying on it as fact.
4. Use appropriate language: ChatGPT is designed to understand and generate natural language, so it’s best to communicate with it in a conversational tone. Avoid using overly complex or technical language that could confuse the model.
5. Be respectful: ChatGPT is a tool designed to assist and provide information, so it’s important to be respectful and use appropriate language when communicating with it. Avoid using offensive or inappropriate language or asking questions that could be considered rude or insensitive.

By following these best practices, you can get the most out of ChatGPT and generate accurate, relevant, and helpful responses to your questions and prompts.”

Chat Entry

“Is there anything ChatGPT is not recommended for?”

Chat Reply

“While ChatGPT is a powerful tool for generating text-based responses, there are some use cases for which it may not be recommended:

1. Critical or sensitive tasks: ChatGPT may not be the best tool for critical or sensitive tasks that require high levels of accuracy or confidentiality. This could include tasks such as medical diagnoses, legal advice, or financial planning.
2. Emotional support: While ChatGPT can provide helpful information and guidance, it is not a substitute for emotional support from trained professionals. If you are struggling with mental health or emotional issues, it’s important to seek help from a licensed therapist or counselor.
3. Real-time conversations: While ChatGPT can generate responses quickly, it may not be the best tool for real-time conversations that require immediate feedback and interaction. In these cases, it may be more effective to communicate with a human through a chat or messaging platform.
4. Creating original content: While ChatGPT can generate text-based responses, it may not be the best tool for creating original content such as articles or reports. The model is designed to generate responses based on existing data and may not be able to create truly original content.

By keeping these limitations in mind, you can use ChatGPT effectively and appropriately for your needs.”

Chat Entry

“Are there other limitations?”

Chat Reply

“Yes, there are some other limitations to keep in mind when using ChatGPT:

1. Lack of common sense: While ChatGPT can generate text-based responses based on its training data, it does not have a true understanding of the world and lacks common sense. This means that it may provide responses that are technically correct but don’t make sense in the context of the real world.
2. Biases in the training data: Like any machine learning model, ChatGPT is only as good as the data it is trained on. If the training data contains biases or inaccuracies, these biases may be reflected in the responses generated by the model.
3. Inability to understand images or audio: ChatGPT is designed to process and generate text-based responses, so it is not able to understand or process images, audio, or other non-textual data.
4. Limited knowledge scope: While ChatGPT has access to a vast amount of knowledge, it is still limited by the scope of its training data. It may not be able to answer questions outside of its domain of knowledge or generate responses to highly specialized or niche topics.

By keeping these limitations in mind, you can use ChatGPT effectively and understand its capabilities and potential drawbacks.”

Chat Entry

“Can you cite sources for this information?”

Chat Reply

“Certainly! Here are some sources to support the information provided:

1. Best practices for using ChatGPT:

• “5 Tips for Using GPT-3 in Your Business” by Mark Armstrong on VentureBeat: https://venturebeat.com/2020/08/08/5-tips-for-using-gpt-3-in-your-business/

2. Limitations of ChatGPT:

• “What GPT-3 can do and can’t do” by OpenAI: https://openai.com/blog/what-gpt-3-can-do-easily-and-what-it-struggles-with/
• “GPT-3: An AI Breakthrough, but not Coming for Your Job” by The New York Times: https://www.nytimes.com/2020/07/29/technology/gpt-3-artificial-intelligence-language.html

I hope this helps!”

[ChatGPT URL used for this document: https://chat.openai.com/chat/70becc05-32f9-48ff-b8c7-26082af048b8]