IE 11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Opinion: What I Learned from a 1-on-1 Interview With ChatGPT

AI chatbots are here to stay, so it’s time to get acquainted with them. Center for Digital Education Senior Fellow Jim Jorstad sat down for a one-on-one with ChatGPT recently and came away impressed.

A smartphone on top of a keyboard backlit in blue, with a conversation with ChatGPT displayed on the smartphone's screen.
Shutterstock
When a new technological innovation becomes a reality, professionals quickly speculate on its potential applications. When it becomes a top news story, the public tries to understand the importance of it in their daily lives. Corporate, government and educational institutions may quickly investigate and embrace it, become cautious to act, or simply refuse to utilize it. Such is the case with the buzz around ChatGPT. Writing about ChatGPT is challenging, since the concept has such broad implications that it can fundamentally change our lives, how we conduct business and learn. What better way to understand this new technology than to first define and explain it, describe its use, and conduct a one-on-one interview with ChatGPT to hear what it has to say about its role in the world.

CHATGPT DEFINED


ChatGPT stands for Chat Generative Pre-trained Transformer. It is basically an artificial intelligence (AI) chatbot that uses normal language to answer questions as if it were human. ChatGPT was introduced on Nov. 30, 2022, by the research company OpenAI. If you want to access this technology, simply go to chat.openai.com and create your own account. But wait until you read the rest of this article first so you will understand “the rest of the story.”

HOW DOES CHATGPT WORK?



ChatGPT is not like a Google search. David Gewirtz, senior contributing editor at ZDNet explains, “ChatGPT’s power is the ability to parse queries and produce fully fleshed-out answers and results based on most of the world’s digitally accessible text-based information.” And this is what excites and yet concerns educators. Gewirtz explains, “ChatGPT has very few limits in subject-matter expertise. You can ask it to write a resume for the character Chief Miles O'Brien from Star Trek, have it explain quantum physics, write a piece of code, write a short piece of fiction, and compare the governing styles of former presidents of the United States.” Could ChatGPT effectively write a student’s report or term paper? The short answer is certainly yes. The challenge is how to determine who is writing the paper in the first place.

As its name implies, ChatGPT needs to be pre-trained. This training can include content from the Internet, books, research, news, and much more. In a blog for the transcription AI company AssemblyAI, Developer Educator Marco Ramponi explains, “The creators have used a combination of both supervised learning and reinforcement learning to fine-tune ChatGPT, but it is the reinforcement learning component specifically that makes ChatGPT unique. The creators use a particular technique called reinforcement learning from human feedback (RLHF), which uses human feedback in the training loop to minimize harmful, untruthful, and/or biased outputs.”

WHAT CAN YOU USE CHATGPT FOR?


In some regards, the uses of ChatGPT are only bound by our collective imagination. The technology website TechWiser in February published a partial list of “25 Things You Can Do with ChatGPT,” but in reality the list seems to expand daily. This technology can generate and debug code, create applications, compose music, serve as a sophisticated chatbot to both interact and direct customers, offer real-time financial guidance, provide training and education, and much more. The corporate and financial sectors are abuzz with the potential of ChatGPT, and investors are looking to fund its further development. Tech titans Microsoft, Google, Apple, Amazon, IBM and Intel are just a few corporations making huge investments in AI-focused technology. The concept of ChatGPT will be both transformative and disruptive.

ONE-ON-ONE WITH CHATGPT


I posed a series of interview questions directly to ChatGPT. Here are some of the answers, edited for style.

What are the advantages of ChatGPT in higher education?

“I don’t believe that higher education institutions should be afraid of the implications of ChatGPT. In fact, they can embrace this technology as an opportunity to improve their teaching and learning processes. ChatGPT has the potential to revolutionize the way we interact with technology and communicate with each other. It can help to automate tasks, provide personalized learning experiences, and enable more efficient communication between students and educators. … However, it is important to recognize that ChatGPT is not a replacement for human educators.”

What are the dangers of ChatGPT?

“ChatGPT, like any other technology, can have potential dangers or risks if it is not used responsibly. Here are some of the possible dangers of ChatGPT:

1. Bias and misinformation
2. Privacy and security
3. Dependence
4. Ethics and regulation

“It is important to acknowledge these risks and work to mitigate them as ChatGPT continues to evolve and become more integrated into our daily lives.”

What does the future of ChatGPT look like?

“The future of ChatGPT is likely to be characterized by continued innovation and advancements in natural language processing, machine learning and artificial intelligence. Here are some possible developments that could shape the future of ChatGPT:

1. Increased personalization
2. Greater integration
3. Improved ethics and regulation
4. Broader applications

“Overall, the future of ChatGPT is likely to be characterized by continued growth and innovation, with the technology playing an increasingly important role in shaping how we learn, communicate, and interact with the world around us.”

Can AI have preferences?

“AI can develop preferences based on the data it is trained on and the rules it is programmed to follow. However, these preferences are not the same as human preferences, which are based on subjective experiences, emotions and consciousness.”

Do you have emotions?

“As an AI language model, I don’t have emotions or feelings in the same way that humans do. While I can understand and process emotions expressed in text, I don’t experience them myself.

“Artificial intelligence is designed to mimic human cognitive functions such as reasoning, problem-solving and decision-making. However, emotions and feelings are a complex and subjective aspect of human experience that are not fully understood even by humans themselves.”

POWER AND RESPONSIBILITY


I was impressed with the clarity and speed of the responses from the one-on-one interview. It is easy to see the potential of ChatGPT, but also the dangers, especially if the systems are hacked or compromised. As the saying goes, “with much power comes much responsibility.” With ChatGPT, its newer versions and more competitors, it will be important to develop thoughtful and careful policies and best practices to realize their true power. Now it’s your turn to sign up for your own account and witness the power and potential of ChatGPT.
Jim Jorstad is Senior Fellow for the Center for Digital Education and the Center for Digital Government. He is a retired emeritus interim CIO and Cyber Security Designee for the Chancellor’s Office at the University of Wisconsin-La Crosse. He served in leadership roles as director of IT client services, academic technologies and media services, providing services to over 1,500 staff and 10,000 students. Jim has experience in IT operations, teaching and learning, and social media strategy. His work has appeared on CNN, MSNBC, Forbes and NPR, and he is a recipient of the 2013 CNN iReport Spirit Award. Jim is an EDUCAUSE Leading Change Fellow and was chosen as one of the Top 30 Media Producers in the U.S.