Chatting with ChatGPT

For this week’s assignment, I chose to do option 1. I’ve never worked with ChatGPT before, and I wanted to know if it works. The first thing I had to do before using ChatGPT was to create an account. After I created my account, it gave a brief overview of how the AI works and all information it holds. It popped up a message saying it collects data and mentioned that I shouldn’t share any sensitive information in my conversations. Later, it started by showing some example questions, the capabilities it holds, and the limitations it has. This helped me because I was confused about how to use AI and what questions it could answer.

To start off, I didn’t really know which question to ask, so I used one of the example questions that ChatGPT suggested. The question that I chose was ‘How do I make an HTTP request in Javascript?’ I thought this was a good question because I knew the answer to that question, and I could cross-check how accurate the answers would be. It in fact did give me the right answer, which is “you can make an HTTP request in javascript using the ‘XMLHttpRequest’ object or the more modern ‘fetch‘ API.” It later gave me 2 examples, one for each choice, which was also accurate. I think in this case the API knew that I was asking a technical question, and it answered the questions perfectly, nothing less or nothing more.

Secondly, I asked ChatGPT what would be a good gift for my valentine. It answered by saying, there are multiple things you could gift for your valentine, and the choice would vary from different preferences. It also gave me 6 different choices, which were really good ones and something that someone would find helpful. At the very end, it also added that the most important aspect of any gift is that it comes from the heart and reflects the love and affection you have for your Valentine. I liked how it was trying to communicate with us as a friend and how it suggested different things to help us. Unlike the first case, for the questions, ChatGPT answered like a friend and was considering human emotions.

Lastly, I asked ChatGPT what would be a better major to choose, Psychology or Computer Science? It answered by telling that it would depend on one’s personal interests and what they’re more passionate about. It listed the facts about each major and what you could consider before choosing either one. For Psychology, it said that If you enjoy understanding human behavior, emotions, and mental processes, and are interested in helping others through therapy or counseling, then psychology might be the right major for you. Whereas, for Computer Science, it stated that if you have a passion for technology, problem-solving, and using computational tools to create innovative solutions, then computer science might be a better fit. I agree with what ChatGPT said because it would totally depends on what one person prefers and feels passionate about. The reasons that ChatGPT suggested were also reasonable and well-stated. It concluded by saying, “Ultimately, the best major for you is the one that you’ll enjoy studying and that will give you the skills and knowledge you need to succeed in the career you want to pursue.” Again like the second case, ChatGPT answered this question in a well-mannered way and sounded in a very friendly tone.

Considering the responses to the questions that I asked, I don’t think there were many limitations in the API. The answers that it gave were reasonable, well-written, and also understandable. I think I would use the AI model languages more often in the future, but it would ultimately depend on the way I want to use them. The most reasonable way to use it would be to ask questions that your peers can’t answer, or if you need help with the work you’re doing. The ethical and social implications that the AI would have could be that they might not tell the right answer when it comes to an emotionally dependent question.

To conclude, I would say that ChatGPT has answers to different types of questions and it answers them according to the question that we ask. I think it would be of good use if we use it for the right purposes and reasons.