Maximizing learning productivity with ChatGPT: What NOT to do
Discover the future of learning as ChatGPT transforms the education experience. Beware of the pitfalls.
ChatGPT has been all the rage since its launch. The idea that ChatGPT can pass professional exams and be a doctor or lawyer is very misleading. It is simply a language model that uses deep learning algorithms to generate human-like text responses based on the input it receives via a prompt. ChatGPT has been trained on large volumes of text data, making it capable of generating natural language responses to a wide range of questions using prompts.
I am an educator teaching my students how to use ChatGPT ethically, and responsibly, to maximize their productivity, yet one can easily fall into bad habits very quickly. As educators, it is critical that the following seven steps are taken into consideration when we use ChapGPT or when our students are using it. This article also provides some guidance on how to avoid plagiarism or AI plagiarism.
7 critical mistakes to avoid when using ChatGPT
- Relying solely on ChatGPT for fact-checking or research
While ChatGPT can provide general information, it is important to verify the accuracy of its responses through additional sources. There are 2 simple ways to do this verify the facts via research articles references or credible online sources or ask chatGPT to provide only credible sources with a reference. Both are equally effective.
2. Quoting ChatGPT as an expert
ChatGPT is an AI language model, not a subject matter expert. Do not quote its responses as if they are coming from an authoritative source. We don't allow students to quote Wikipedia, so don't make them quote ChatGPT, and neither should you.
3. Expecting consistent answers
ChatGPT is a probabilistic model, meaning that it generates responses based on the likelihood of certain words or phrases appearing together. This means that its responses may vary, even for the same user prompt. This breaks the definition of insanity, doing the same thing can give you something different, so please don't expect a student Prompt to generate what you are expecting to see.
4. Expecting it to perform tasks outside of its capabilities
ChatGPT is not capable of performing tasks such as writing code, making financial predictions, or diagnosing medical conditions. However, with the correct prompt and context you can perform these tasks that support some of these high-level expectations. If you intend to perform high-level tasks, you need to master prompt engineering first.
5. Expecting perfect grammar and spelling
ChatGPT was trained on a large corpus of text, but it is not immune to making grammatical or spelling errors. It is important to double-check its responses before using them for important tasks. Use the Grammarly app if you are not an English major
6. Overreliance is Precarious
While ChatGPT can be a valuable tool, it is not a replacement for human judgment and critical thinking. Use it as a resource to support your decision-making, but don’t rely on it as your sole source of information or guidance.
7. Plain Plagiarism or AI Plagiarism
In the academic context this is probably the most important consideration, so let's spend a min to cover this in depth. GPTZero is a program that helps determine if the text was created by AI or you. Plagiarism checkers help to ensure that the content is original and not copied from other sources. The example below provides some guidance on how to verify your generated output will pass.
How to avoid AI impersonation fraud” or “AI plagiarism.”
Avoiding AI plagiarism is particularly important in academic and professional settings, where presenting plagiarized or misrepresented work can have serious consequences, such as a zero on an assignment, loss of credibility, and potential legal action. In these settings, it is crucial to give credit to the original authors and ensure that the work being submitted is not inadvertently copied from someone else’s work. Let's review the example text below.
GPTZero uses two indicators to determine if an excerpt was written by a bot or not. One indicator measures the complexity of the text, while the other compares the variations in sentences. If the text has a high level of complexity, it is more likely to be written by a human, but if the text is familiar to the bot due to its training, it is more likely to be generated by AI. In terms of sentence variations, human-written text tends to have more variation, with a mix of longer and shorter sentences, while AI-generated text is often more uniform.
Was the above paragraph created by ChatGPT or not?
The text passed the plagiarism checker and did pretty well with spelling and grammar, punctuation, conciseness, and readability. Now lets check for AI plagiarism with GPTZero, because humans deserve the truth.
Even though the paragraph will pass the ZeroGPT test, It is really important to address any issues identified by the tool. Hence the two sentences must be reworded.
Now for the big reveal. The paragraph was created by ChatGPT with some clever prompt engineering.
ChatGPT Prompt: Can you rewrite this statement to avoid “perplexity” and “burstiness indicators. To determine whether an excerpt is written by a bot, GPTZero uses two indicators: “perplexity” and “burstiness.” Perplexity measures the complexity of text; if GPTZero is perplexed by the text, then it has a high complexity and it’s more likely to be human-written. However, if the text is more familiar to the bot — because it’s been trained on such data — then it will have low complexity and therefore is more likely to be AI-generated. Separately, burstiness compares the variations of sentences. Humans tend to write with greater burstiness, for example, with some longer or complex sentences alongside shorter ones. AI sentences tend to be more uniform.
Thanks for sharing the unfolding journey of how AI will change education. I will be creating some more posts on this topic. I hope you found the information useful and informative please don’t hesitate to clap, and follow me. If there’s anything you’re curious about reach out and ask!