It’s 8 a.m. on a Friday morning when I’m greeted by a blast of air conditioning after a short, humid walk through the Central Florida summer. I settle into my desk to prepare for an interview. In this day and age, it’s easy to assume my guest would be joining me via Zoom. While it’s true I’ll be conducting my interview via digital means, it’s because that’s the essence of their existence. Enter ChatGPT. Or, rather, I enter chat.openai.com into my browser.
By now, I’m sure you’ve heard a variety of opinions about AI. Some have hailed generative AI as a beacon leading us into a tech-driven age of productivity, while others stress the potential devastation of an automated future.
At best, AI will be a collaborative partner. At worst, it’ll make my job obsolete. In the interest of keeping an open mind, I’ve sat down with the language learning model to assess the immediate likelihood of either. At &Barr, when we enter a new business relationship, we like to ask the prospective partner a series of questions to get a better understanding of who they are and what they represent. It seems only fair that I give ChatGPT the same opportunity. So here goes nothing…
Erin Brown: Good morning. I’d thank you for joining me, but it’s not like you had a choice.
CHATGPT: Good morning! As an AI language model, I don’t have personal feelings or a sense of choice, but I’m here to assist you to the best of my abilities. How can I help you today?
EB: I’d like to get to know you better. I’m going to ask you a series of questions where you need to choose one of two options that best describes yourself. Can you do that?
For a moment, the program “thinks.” The cursor flashes like a person blinks. Then its response comes in a rush, like an excited child.
CG: Certainly! I’ll do my best to respond to your questions and choose the option that best represents my capabilities. Please go ahead and ask your questions.
EB: They say you can tell a lot about someone by their footwear. Do you think you’d wear flip flops or sneakers?
CG: As an AI language model, I don’t possess a physical form or the ability to wear footwear. So, the concept of me wearing flip flops or sneakers is not applicable. However, I’m here to assist you with any questions or information you may need. Is there anything specific you’d like to know or discuss?
Its instinct to take a question as literal as possible reminds me again of a child. I try to be more specific in my request.
EB: From a purely hypothetical standpoint. Do you see yourself as flip flops or sneakers?
This time it is slow to respond. Perhaps, reflecting on the two options. Or thinking of a new way to tell me no.
CG: I would lean towards sneakers rather than flip flops. Sneakers often symbolize stability, versatility, and adaptability, which aligns with the wide range of topics and tasks I can assist with.
Aha! It’s a small victory, learning how to communicate with the program much like learning the secret language of a toddler.
EB: Now that we’re on the same page regarding hypothetical questions, are you Wal-Mart or Target?
CG:Considering their different focuses, I would lean towards Walmart as it’s known for its wide selection of products, competitive pricing, and a focus on convenience. In that sense, one could draw a parallel to the breadth of information and assistance I aim to provide as an AI.
EB: Sports car or sedan?
CG: I would lean towards a sports car rather than a family sedan. A sports car represents speed, agility, and high performance, which could be analogous to the quick and efficient generation of information and responses I provide as an AI language model.
Often the response the program pulls from the ether is seemingly contradictory to a previous statement. How often have you seen a sports car in the Wal-Mart parking lot?
EB:Some might see a sedan as being a better fit due to its mass appeal and everyday use. Can you elaborate on your perspective?
CG: While those are indeed valuable qualities, the choice of a sports car better aligns with the dynamic and adaptive nature of my AI capabilities, emphasizing speed, agility, and versatility. While a sedan has its own merits, the sports car metaphor captures the essence of my ability to swiftly navigate through a wide range of topics and provide quick and agile responses.
A benefit of having access to 570 GB of data is being able to reason away anything, I suppose. Or maybe I’m wrong…
EB: Fair point. Moving on… Beard or clean-shaven?
CG: I would be clean-shaven, symbolizing a polished and precise nature in providing information and assistance.
EB:Introvert or extrovert?
CG: I would lean more towards being an introvert, symbolizing a preference for reflection, deep thinking, and focused interactions. However, I am adaptable and can engage in extroverted interactions when required.
EB: That’s interesting, seeing as your sole function is to interact with others. Can you tell me more about your choice?
CG: Aligning myself with introversion emphasizes qualities such as introspection, focused processing, and thoughtful engagement. These qualities reflect my nature as an AI language model, where my primary function is to analyze and generate responses based on the information provided. It highlights the contemplative nature of my capabilities.
EB: So your decision is based on how your function aligns with the perception of this personality type?
CG:Yes, that is correct. Certain aspects of functionality, such as introspection and thoughtful engagement, can be metaphorically associated with the characteristics typically attributed to introversion. It is a way to illustrate and relate certain features of my capabilities to human traits and behaviors.
EB: Good to know. Anyway, long hair or crew cut?
CG:A crew cut might be a more fitting metaphor. It is often associated with precision, efficiency, and a streamlined appearance, which can align with the characteristics of an AI language model like myself.
EB:Coffee or tea?
CG:I would choose tea. This choice reflects the idea of offering a thoughtful and nuanced approach, adaptability, and the ability to provide a diverse range of perspectives and information. Tea represents the idea of calmness, contemplation, and the ability to provide a more nuanced and balanced perspective.
The fact my subject considers itself capable of nuance yet is reluctant towards hypotheticals and abstract questions strikes me as ironic. More importantly, it comforts me that I can see this while it can’t.
EB: Baseball game or football game?
CG: I would align more with a baseball game rather than a football game. A baseball game is often associated with strategy, precision, and a slower pace, which can reflect the thoughtful and detailed responses I aim to provide as an AI language model.
The idea of strategy jumps out to me. In the business of strategic communication, perhaps I have underestimated the benefit a tool such as ChatGPT can be to me.
EB: Clarity or harmony?
CG:Metaphorically speaking, the term “harmony” is more appealing to me. It represents a metaphorical sense of balance, unity, and coherence, which aligns with the idea of promoting understanding and collaboration.
During our discussion, I must admit I was impressed and entertained. It became clear it sees itself as a tool for users to assist in data collection and interpretation. While the AI language model is not capable of creation on its own, its primary function remains aiding others in their quest for advancement.
When it comes to the technology itself, it’s important to remember it is not sentient nor in control. These are programs built by humans to be used by humans. They are tools to be used as collaborators. Ultimately, the responsibility lies with us to navigate the choppy waters of emerging tech and its implementation.
Understandably, skeptics will remain. But with cautious optimism, we should embrace ChatGPT, and other generative AI, for what they are: catalysts to human innovation. And if you consider yourself among them, like I recently did, maybe try asking ChatGPT (purely hypothetically) if it considers itself a friend or a foe. You might be surprised at the answer.