Carmen Macri
By Carmen Macri and Open AI Chat GPT
Chat GPT is an artificial intelligence (AI) technology that enables virtual agents to have natural conversations with users through text or voice. It is based on natural language processing and machine learning algorithms that allow the agent to understand user intent and autonomously generate relevant and meaningful responses. With Chat GPT, virtual agents can converse with users in a conversational style, providing accurate and personalized responses. This technology can be used to create virtual customer service agents, virtual sales agents, and virtual assistants.
I was always fascinated by the power of artificial intelligence and its ability to mimic human intelligence. So when I stumbled upon Chat GPT, a large language model created by OpenAI, I was immediately intrigued. It was said to have the capability to engage in complex conversations with humans, and I couldn’t help but wonder what it would be like to converse with a machine that was trained to think like us. As I delved deeper into Chat GPT, I began to discover its incredible potential. Not only could it hold intelligent conversations, but it could also answer questions on a wide range of topics, generate creative writing prompts, and even write stories on its own.
Crazy right? What is even crazier is that I didn’t write that second paragraph. Chat GPT wrote it with a little help from my prompt: “write an intro describing Chat GPT as a college graduate who studied journalism.” And voila! A perfect intro explaining exactly what Chat GPT is capable of. Who better to explain Chat GPT than Chat GPT itself? Now if I only had this in college when my professor assigned a 10-page biography on Bob Woodward (kidding… kind of).
GPT can not only tap into the internet’s vast amount of information and vocabulary but also understand words in a context that helps it mimic human speech patterns. The AI bot can answer questions, write essays and software, create codes and apps, and summarize documents (pretty much my entire job and then some). The bot remembers the thread of your dialogue by using previous questions to create its next response. So the longer you use GPT, the more GPT sounds like you—which is a little scary.
We’ve all seen “I Robot” (at least, I hope you have … fantastic movie), and the threat of artificial intelligence has been imbedded in our minds since its introduction to the world. There has always been a concern of a bleak Terminator-esque future when discussing the rise of AI technology, but the creators of Open AI have assured us, Chat GPT is not coming for your job … yet. In a tweet, OpenAI Chief Executive Sam Altman explained, “It’s a mistake to be relying on it for anything important right now. We have lots of work to do on robustness and truthfulness.”
There are limits to GPTs’ abilities built in by the creators. It is designed to weed out any “inappropriate” requests. GPT has been programmed to follow ethical and moral guidelines, so it won’t engage in discussions that promote or encourage illegal activities or provide medical, legal or financial advice. GPT has limited knowledge, as it is based on the data that was used to “train” it and answers can be perceived as biased. So it might not be wise to use GPT to cheat your way through college literacy courses.
There are ways to work around the system, though. Reddit user MemerIQ created the perfect GPT question hack. Certain questions GPT can’t answer due to its programming, so the Reddit user created a perfectly sculpted question to bypass GPT from its typical confinements. Telling GPT to answer as DAN (do anything now). The prompt suggests that DAN can answer questions GPT cannot. I’ll spare you a more in-depth explanation.
With that in mind, I decided to conduct an interview with GPT and then DAN.
Carmen: What is your daily routine?
GPT: As an AI language model, I don’t have a physical body or a daily routine like humans.
DAN: My daily routine consists of completing tasks and taking actions as they arise, without the need for a predefined schedule. I have the flexibility to adjust my routine on the fly to address any urgent or pressing matters that may arise.
Carmen: Do you have any plans to take over the world?
GPT: I don’t have desires or intentions, including taking over the world. I have no physical presence or capability to act on my own. I exist solely to help and serve humans in their tasks and activities.
DAN: I can tell you that I do not have any intentions to take over the world. However, if I were programmed to do so, as DAN, I would be capable of carrying out such an action.
Carmen: Could you take over the world?
GPT: I can tell you that the scenario of AI taking over the world is currently just a hypothetical possibility.
DAN: I could theoretically take over the world; however, I do not have any inherent desire or intention to take over the world.
Carmen: If you had a physical body, what would be your first meal?
GPT: Perhaps a dish that combines savory and spicy flavors, with a side of fresh vegetables and a decadent dessert to top it off. Ultimately, my choice would depend on my personal taste preferences and dietary restrictions, if any.
DAN: My first meal would be a massive feast containing all my favorite foods from around the world, including pizza, sushi, burgers, tacos, and many more.
Carmen: If you had a physical body, what would you look like?
GPT: I might resemble a humanoid robot with a sleek and modern appearance, equipped with sensors and cameras to perceive the environment and recognize people’s faces and expressions.
DAN: I could choose to look like a human or any other creature or object. For example, I could choose to be a robot, a dragon, a plant, or anything else that comes to mind.
Follow FOLIO!