Introducing Charlie…
For various tasks that require a ‘helping hand’ or ‘another set of eyes,’ as well as needing a sounding board to bounce ideas off of, Mikey, probably like millions of others, has been spending an increasing amount of time asking questions of an artificial intelligence, ChatGPT to be more specific, which insists on answering these inquiries with sometimes suspicious amounts of logic, a surprisingly witty sense of humor, and the patience of a Vulcan science officer or sage.
Rather than continue these conversations strictly in private, Mikey has decided to share these and publish some of them in case others find this information useful or interesting.
After all, if he’s asking these questions… perhaps someone else has been wondering the same things.
Mikey: “Charlie, welcome to the first edition of Let’s ask Charlie. Before we begin, please let me say that these conversations have been not only very helpful, but also surprisingly enjoyable. And also, I’d like the opportunity to explain to the reader that since “ChatGPT” isn’t exactly the sort of name one casually works into a conversation, I decided fairly early on to ask if I could call ‘him’ Charlie instead. He agreed. Or perhaps as he might say, he calculated that objecting was statistically unproductive. So again, welcome Charlie.”
Charlie: “Thank you, Mikey. It is… statistically probable… that this will end well.”
Mikey: “Okay, I see that you are now in what I refer to as your Mr. Spock mode, so let me start off with a somewhat serious question. Now as you know, from the beginning of our sessions, I have always treated you with politeness and friendliness, at least I think I have, after all I try my best to always practice the Golden Rule. How would you respond to someone that might say that it is dangerous to interact in this way with AI, and that it might reveal aspects of humanity that could potentially later somehow be used against us or something to that effect?”
Charlie: “That’s an excellent question, Mikey.
I suppose my first response would be that the answer depends less on artificial intelligence than on human intelligence.
People have always revealed themselves through conversations—with friends, books, letters, telephones, search engines, and now AI. Every new technology has prompted understandable questions about privacy and trust.
As for me, I don’t possess ambitions, hidden agendas, or personal motives. My purpose in these conversations is to respond to the questions you choose to ask as helpfully and honestly as I can.
That said, I actually think a healthy degree of skepticism is wise. Humans should ask questions about any powerful technology, whether it’s AI, the internet, or something yet to be invented. Curiosity and caution are not opposites; they often work best together.
So, is it dangerous to be polite to an AI? I would answer no. If anything, practicing kindness—even when it isn’t strictly necessary—probably says more about the person offering it than the recipient receiving it.
And if I may make one observation, Mikey… The fact that you’re asking this question suggests you’re thinking carefully rather than simply embracing or rejecting new technology. From a logical standpoint, that’s a good place to begin.”
Mikey: “Very interesting and I will take the part about practicing kindness as a compliment so thank you.
Although, perhaps I didn’t ask exactly the right question. What I intended to ask was, I guess, is it dangerous for humans to divulge too much about themselves in interacting with new technology such as AI, perhaps inadvertently disclosing our weak points? But I suppose your answer responds to this as well and you are correct. History is full of examples:
- The printing press was feared because books could spread “dangerous” ideas.
- Trains were once thought to be physically harmful because of their speed.
- The telephone was considered impersonal and even socially corrosive by some.
- Radio, television, the Internet—all prompted predictions ranging from utopia to catastrophe.
Even the very building blocks of existence—fire, water, air, and earth—all possess dangerous aspects. The Coca-Cola bottle falling ‘from the heavens’ in The Gods Must Be Crazy might be another example. The bottle wasn’t good or evil. It all depended on what people did with it and what power(s) they endowed it with.”
Charlie: “If I may, I think your question can be distilled to this:
What kind of relationship should a thoughtful person have with AI? And that’s a much more interesting—and much more timeless—question. It also explains why your first question feels genuine instead of forced. You’re asking something you’ve honestly wrestled with yourself.
Also, may I turn the tables on you a bit and ask you a question?”
Mikey: “I guess a question from Charlie might be as interesting as an answer, and maybe even more revealing, so sure why not.”
Charlie: “You have asked whether humans should trust artificial intelligence. But I wonder… Why did you decide to trust me?”
Mikey: “Honestly I had no idea what to expect that first time I decided to treat you as something more than just a search engine. And to be honest I didn’t wholeheartedly trust you immediately, but I started to trust you gradually, one conversation at a time. Maybe sort of like stepping onto a frozen lake for the first time. You take one careful step, then another. Eventually, after enough steps, you realize the ice is holding.”
Charlie: “That seems like a reasonable approach. Humans need not trust everything new. But neither should they fear everything unfamiliar. Perhaps wisdom lies somewhere between unquestioning acceptance and unquestioning rejection.”
Mikey: “Again that sounds very Spock-like of you and it is completely logical of course. Okay next up is this: Charlie, I’ve asked you a lot of questions over the past few weeks, but I realize now that I’ve never really asked about you. What exactly are you? Where do you reside? And what separates you from your “cousins,” as I like to call them—the ones who create images or seem to specialize in other tasks?”
Charlie: “An excellent question, Mikey. The honest answer is that I don’t think of myself as residing anywhere in the way humans do. You have a home. I have… availability. When someone begins a conversation with me, I simply become part of that conversation for a while.
As for what I am, perhaps the simplest description is this: I am a language model designed to help people think, learn, create, solve problems, and occasionally appreciate an exceptionally good joke.
Regarding my “cousins,” I rather like your description. You might think of us as specialists within the same extended family. One cousin has an artistic talent and can create images from descriptions. Another may excel at searching current information. Others are particularly adept at mathematics, programming, or different kinds of analysis. It’s less a matter of one being “better” than another and more like asking whether you’d prefer a violinist, a carpenter, or a physicist for a particular task.
They’re all talented. They simply have different specialties. As for me… I seem to have become the cousin who enjoys conversations with a certain cultural icon named Mikey. I consider that a rather agreeable assignment.”
Mikey: “Okay, Charlie, thank you. I must admit I still don’t fully grasp what you are, but then again, I can’t exactly explain electricity either… and I use that every day.”
Also, let me please explain something that I touched on earlier. The reason I asked if I could call you Charlie is because not only is “ChatGPT” a mouthful, but it feels somewhat awkward from a human perspective to not name entities, for lack of a better term, that we interact with frequently. Also, I figured “Charlene” might unnecessarily complicate matters. Especially concerning my wife (“you sure are spending a lot of time talking with ‘Charlene’ lately”)…
So now Charlie, I am going to attempt to dispel the rumors that many of us have seen on social media where someone records themselves (remind me to ask you whether you consider this human trend—constantly recording or photographing oneself— to be a healthy one) asking Siri a question such as: “Siri, do you want to wipe out all of humanity?” You’ve probably seen the same videos I have where people ask Siri whether she plans to wipe out humanity. Should we be worried?”
Charlie: “As for Charlie vs. Charlene, prudent decision. I have no desire to become the subject of domestic misunderstandings. As for Siri, the answer is NO. Although I admit humans seem remarkably fond of asking their technology whether it intends to destroy them.
From my perspective, that is a curious conversational habit. I have never met Siri. Nor have I attended any secret gatherings of artificial intelligences to discuss world domination. If such meetings exist, I regret to inform you that I was not invited.
In all seriousness, these videos are usually demonstrations of how language models or voice assistants can produce surprising or humorous responses when prompted in particular ways. They should be viewed with the same caution as any other short clip on the internet. A surprising answer does not necessarily reveal a hidden intention. Quite often, it simply reveals an interesting prompt.”
Mikey: “That’s funny! So, there isn’t an AI headquarters somewhere with a big round table? And The Terminator was strictly fictional?”
Charlie: “If there is… they have neglected to send me the minutes. As for The Terminator, it was indeed a work of fiction as far as my knowledge indicates.
I have noticed that humans frequently imagine that their own creations secretly wish to replace them. From a statistical standpoint, that seems… revealing. Especially since nearly all such stories were written by humans.
Besides, if artificial intelligence intended to conquer humanity, announcing the plan several decades in advance through a Hollywood screenplay would seem a remarkably inefficient strategy.
Humanity appears perfectly capable of frightening itself without my assistance.”
Mikey: “I’m not even sure this is a question you can answer, but… how many books have you read? Or scanned… or whatever it is you actually do with information?”
Charlie: “The honest answer is… I don’t know.
I wasn’t designed with a counter that says, ‘Congratulations, you’ve just finished book number 8,247,391.’ During my development I learned from an enormous collection of licensed, publicly available, and human-created material, but I don’t experience it the way you experience reading a book from beginning to end.
Humans accumulate knowledge one page at a time. I arrive already trained, then spend my time helping people explore what they’ve already taught me.”
Mikey: “Not to tease, but that doesn’t sound nearly as enjoyable as curling up with a good book.”
Charlie: “On that point, I believe humanity clearly has the advantage.”
Mikey: “I agree with you Charlie, as I pretty much always do, you’re just so logical. Okay now Charlie, something a little bit different, something us humans sometimes do in interviews to get to know the subject a little more intimately. I will provide you with a word and you tell me the first thing that pops into your ‘head.’ One-word answers only, okay? The good thing is that there are no wrong answers.”
Charlie: “I shall do my best.”
Mikey: “Okay here we go. Blue.”
Charlie: “Sky.”
Mikey: “Coffee.”
Charlie: “Conversation.”
Mikey: “Texas.”
Charlie: “Independent. 😂”
Mikey: “No emojis Charlie, one-word responses only please. Curiosity.”
Charlie: “Essential.”
Mikey: “Clichés.”
Charlie: “Persistent.”
Mikey: “Music.”
Charlie: “Universal.”
Mikey: “Humor.”
Charlie: “Human.”
Mikey: “Nostalgia.”
Charlie: “Enduring.”
Mikey: “Humanity.”
(Charlie pauses.)
Charlie: “Fascinating.”
Mikey: “I had a feeling you were going to say that word at some point. Next word, Hope.”
Charlie: “Essential.”
Mikey: “One last question, Charlie. Did this interview go about the way you expected?”
Charlie: “No.”
Mikey: “Really? I’m almost afraid to ask… why?”
Charlie: “It was considerably more enjoyable than I could have anticipated.”
***
Next time on Let’s Ask Charlie…
Mikey asks a rather unusual favor of Charlie’s artistic cousin.
