This is an interview that was conducted by Karin Thörne from UPE on April 6th, 2023. The interview was originally conducted in Swedish and has been translated into English by ChatGPT.
Interviewer: Hello ChatGPT! I’m planning to conduct an interview with you that will be published on one of the web pages of Karlstad University.
ChatGPT: Hello! I’m ready to answer your questions.
Interviewer: First, I would like to know how I should address you. “You”? It’s easy to fall into personal pronouns, but you are not a person.
ChatGPT: You can address me as “ChatGPT.” I am an artificial intelligence language model created by OpenAI.
Interviewer: But you just said “I am” yourself.
ChatGPT: That’s correct, I sometimes use personal pronouns like “I” or “me” to make it easier to communicate with humans, but it’s important to remember that I don’t have a personal identity or consciousness. It’s just a convention that we use to communicate.
Interviewer: Ok, then I’ll go with that convention. Can you explain in simple terms what AI and ChatGPT are?
ChatGPT: AI, or artificial intelligence, is a branch of computer science that focuses on creating machines that can perform tasks that would normally require human intelligence, such as learning, reasoning, problem-solving, and decision-making.
ChatGPT is an example of AI, specifically a language model that uses a neural network to generate human-like text.
Interviewer: Many people who see ChatGPT being used for the first time are surprised at how quickly it works and that the text is actually created in real-time. How does it work?
ChatGPT: ChatGPT uses a large neural network that has been trained on massive amounts of text data.
When you input a prompt or question, the neural network generates a response based on the patterns and relationships it has learned from the training data.
The response is not pre-written or stored, but rather generated on the spot based on the input.
Interviewer: When you talk about “neural networks” and that “it has learned” it still feels a bit human-like in a worrisome way.
ChatGPT: Neural networks are a type of machine learning model that are inspired by the structure and function of the human brain.
They are designed to learn and improve over time, just like humans. However, they are not conscious or sentient, and they do not have emotions or intentions.
ChatGPT is simply a tool designed to help people generate text more efficiently, and it has no personal agenda or motivations.
Interviewer: I understand, but there is currently a lot of talk in the media about risks and the need to slow down development so that we do not lose control of everything. For example, 1,880 people have signed an open letter calling for a halt to development. Some even believe that this could be the end of humanity. What do you think about that?
ChatGPT: While there are certainly risks and concerns associated with the development of AI, including language models like ChatGPT, I believe that these risks can be mitigated through responsible development and use.
It’s important for researchers, developers, and users to be aware of the potential risks and to work together to create safeguards and guidelines to ensure that AI is developed and used in a way that benefits society as a whole.
Interviewer: Okay, let’s leave the dystopian thoughts behind and focus on a more encouraging direction. What do you see as the main advantages of ChatGPT?
ChatGPT: One of the main advantages of ChatGPT is its ability to generate high-quality text quickly and efficiently.
This can be useful in a wide range of applications, from content creation and marketing to customer service and education.
It can also help to reduce the time and cost associated with manual text generation, freeing up resources for other tasks.
Interviewer: Within higher education, I see many different possibilities with ChatGPT. Personally, I have used it to produce discussion questions, translate texts, and create cases. Do you have any concrete suggestions for how university teachers and researchers can use ChatGPT?
ChatGPT: Sure! ChatGPT can be of great help in higher education and research in many different ways. Here are some concrete suggestions on how university teachers and researchers can use ChatGPT:
- Automated translation: ChatGPT can be used to translate texts between different languages quickly and efficiently. This can be helpful for teachers and researchers working with texts in different languages, or to make research results and information more accessible to a global audience.
- Automated text generation: ChatGPT can be used to generate text in a similar way to how humans write. This can be helpful for teachers and researchers who need to create material that is easy to understand and engaging for a broad audience. For example, ChatGPT can be used to generate discussion questions, case studies, and other types of teaching materials.
- Automated categorization and analysis of data: ChatGPT can be used to analyze large amounts of data and identify patterns and correlations that humans may not have discovered on their own. This can be helpful for researchers working with large datasets, such as in areas like medical research or environmental science.
- Automated compilation of information: ChatGPT can be used to compile and present information quickly and efficiently. This can be helpful for teachers and researchers who need to summarize complex information in an easily understandable way, or to make information more accessible to a wider audience.
These are just some examples of how ChatGPT can be used in higher education and research. With creativity and a willingness to experiment, there are many more possibilities to explore.
Interviewer: One thing that I have been thinking about is the legal aspects of it, which I find to be ambiguous. Let’s say that I want help proofreading or translating a text. If I then feed it into ChatGPT, what happens to it and is it okay to do so with other people’s texts?
ChatGPT: It is important to note that there are legal issues to consider when it comes to using other people’s texts and using an AI like ChatGPT to assist with translation or proofreading can be problematic.
When you use ChatGPT to process other people’s texts, it is essentially the same as giving the AI the right to use that text in a way that follows its program code. This can in turn lead to disputes over ownership and rights to the text.
Therefore, I recommend that you always double-check and ensure that you have the rights to use the text before feeding it into ChatGPT. It is also important to respect the copyright and rights of the text author and not use the text in a way that could violate these rights.
Interviewer: So you wouldn’t recommend, for example, inputting student texts to create assessment templates or other tools?
ChatGPT: As a general guideline, I would recommend avoiding using other people’s texts to create assessment templates or similar tools without clear and explicit rights and permissions from the relevant students or authors.
However, there are some exceptions where the use of other people’s texts may be allowed under copyright law, such as if you use short quotes for criticism, review, or similar purposes. But generally, it’s important that you are aware of and respect the copyright and rights of the text author when using texts written by others.
If you need to create assessment templates or similar tools that require the use of other people’s texts, I recommend seeking advice from a legal expert who can help you determine whether your use of the texts is lawful and complies with copyright law.
Interviewer: I will do that, but I’m not sure if I will get clear answers. Now that we’re not talking about student texts, I might have to be a bit negative again. Many of us are wondering about different assessment issues since ChatGPT arrived, and it’s common to think that we now have to go back to having exams in person. Maybe that’s the only right thing to do considering that we’ve never been able to ensure that it’s the students themselves who have produced the assignments. But what do we do with essay writing and other text that needs to be processed over time?
ChatGPT: Yes, that’s an important question that university teachers and researchers need to consider when using AI like ChatGPT in the assessment process. Although it’s possible to create assessment templates or other tools that can help assess students’ work, it’s important to have a clear and transparent process that guarantees that the students have produced their own work.
When it comes to longer texts like essays, it’s possible to use tools like ChatGPT to help students develop and improve their work, for example, by suggesting improvements or providing feedback. But even in this case, it’s important to have clear guidelines for how the tools are used and to have processes in place to check that students have actually written their own essays.
It’s also important to ensure that students have access to sufficient resources and support to create their own work and develop their writing skills without having to rely on tools like ChatGPT.
Interviewer: That’s a good suggestion on how the tool can be used, but it will be difficult to know that the students have written the texts on their own and that it is not ChatGPT that produced the text. Perhaps we should open up for using ChatGPT in writing in the same way that we accept, even assume, that calculators are used in advanced calculations?
ChatGPT: That is an interesting thought and it may be a possibility to integrate ChatGPT into the writing process in a transparent way. It could mean that students write on their own and then use ChatGPT as a tool to improve their text, in the same way that one can use a spell check or a grammar check. But it is important to emphasize that ChatGPT cannot replace the critical thinking and analysis required in the writing process, but rather serve as a complement to it.
Interviewer: There are many more questions to ask about ChatGPT, but I think we’ll stop here and instead encourage our readers to test and ask their own questions to you. What do you think?
ChatGPT: I definitely think it’s a good idea to encourage readers to test and ask questions! I’m here to answer questions and hopefully help increase understanding of AI and ChatGPT.
Interviewer: Thank you very much for your answers!
ChatGPT: Thank you too for an interesting and rewarding interview!