Saying "Please" to the Artificial Intelligence
I recall when it first happened. It was about four years ago. I had just gotten my first pair of wireless earbuds and was heading out for a run (that’s how you know it was a while ago). Gliding across the pavement, I double tapped a bud to activate Siri–the friendly artificial intelligence incarnated by Apple–in order to request that she increase the volume of my playlist. The road noise was killing my vibe.
And that’s when I thought, “Wait. Do I say please?” As in, “Siri, please turn up the music’s volume.” Should you say please to an AI?
—
The 18th century German philosopher, Immanuel Kant, is known as one of the foremost philosophers of ethics. When it comes to animals, Kant says, it’s not so much that we have a duty to treat animals well. For Kant, animals have no innate rights. Rather, we should treat animals well because, in doing so, we develop the kind of internal dispositions that allow us to treat other human beings well. In Kant’s view, the boy who treats his childhood collie with kindness is more likely to become a grown man who treats his spouse with affection.
And so, getting ready to prompt Siri to turn up the Ed Sheeran song, Kant’s stance came to mind, “Siri doesn’t care if I say ‘please’. Not being sentient and all, Siri wouldn’t even mind if I laced my request with demanding expletives. But, what would Kant do?”
Kant, I’m pretty sure, would say please.
—
I harbor some fears regarding artificial intelligence that seem common to folks who have spent too much time watching science fiction movies. Will the AI be good guys or bad guys? Will the AI take all of our jobs? Will the AI make us all lazy and complacent until we are unthinking consumers like those folks from Wall-E?
But, there is deeper stuff, here, too. At the core of it are questions around what it means to be conscious. What it means to be a creature that can think about its own thinking. What it means to be human.
In 2013, the movie Her came out. It features a man (Joaquin Phoenix) who falls in love with an artificial intelligence (voiced by Scarlett Johansson). I haven’t watched the film for about ten years, but from what I recall, a central question raised in the movie is this: Can a human being have a real relationship with an artificial intelligence?
We seem to develop relationships all the time with things that we wouldn’t classify as humans. Our pets, for one. Maybe an old tree growing in the yard. And even our machines–doesn’t a car, after you have driven it tens of thousands of miles, almost feel like an extension of oneself?
So, what about these AI? Can interactions with an AI, such as the chatbot known as ChatGPT, be considered authentic interactions? And what would it take for us to consider an artificial intelligence to be sentient? If a machine passes the classic Turing Test, is that enough for us to say that we have a new sentients on our hands? Can we really reproduce, with wires, electricity, and algorithms, the kind of cognition and metacognition that heretofore we have only discovered in the flesh and bones of ourselves?
—
I recently finished listening to the audiobook, Co-Intelligence: Living and Working with AI, by Ethan Mollick. Being a reflexive skeptic of artificial intelligence and a teacher who is worried about the impacts AI could have on education, I feel that I should be open to learning more about the technology. Hence, I gave this book a listen.
I found Mollick’s takes to be refreshing. He seems to be neither a doomsdayer nor a pollyanna when it comes to artificial intelligence. Mollick makes it clear that we must ensure that the aims of artificial intelligence are aligned with the aims of humanity, while providing helpful suggestions for how we can use AI productively.
Mollick suggests that we are still a ways off from what is known as Artificial General Intelligence, that is, an AI that is capable of novel and responsive thinking–an entity that we might truly consider to have consciousness. That said, Mollick is quick to remind us that we must assume that the AI we are using, today, will be the worst AI that we will ever use. AI will only get better.
As an educator, I alternate between worrying about what AI could mean for learning and being excited at the possibilities. Will students use AI to simply write their entire essays, copy and paste? Or can students be encouraged to use AI mindfully, use it to help create outlines and to brainstorm?
If AI is doing all the hard work for us, will we ever be spurred to have a creative thought, again? Will our ability for novel thinking atrophy as we look to the algorithm for all our inspiration? Will an AI age ever produce the likes of Copernicus, or da Vinci, or Marie Curie, or Einstein, or MLK?
—
Interacting with AI like chatbots sometimes gives me the heebie-jeebies. When I first learned for ChatGPT, I was reluctant to even go to its URL. It felt like, if I began interacting with this AI–training it on my questions, letting it have a peek inside my brain as it began to make a model of me–I was somehow crossing a boundary. A step into dystopia. An experiment with the darkside.
But, that is all a little dramatic. And, I’ve come to grapple with the reality that, if I wish to really understand the possible perils of AI and its potential benefits, I should probably have at least a working familiarity with it.
In that spirit, why not just ask the AI (in this case, ChatGPT) if I should say “please”?
Here is my exchange with the AI.
Me:
Hi, ChatGPT. I have been having a debate with myself about whether, when making a request from an artificial intelligence, should I say "please" to that artificial intelligence. You are an artificial intelligence. What do you think?
ChatGPT:
Whether or not to say "please" to an artificial intelligence is ultimately a matter of personal preference and philosophy. Here are a few considerations:
Reasons to Say "Please":
—Politeness Habit: Saying "please" can reinforce good manners and courtesy in your interactions with others, including AI.
—Human-Like Interaction: It can make the interaction feel more natural and human-like, which some people find comforting.
—Positive Atmosphere: Polite language can contribute to a more positive and respectful atmosphere, even in human-AI interactions.
Reasons It Might Not Be Necessary:
—Functionality: AI does not have feelings or the need for social niceties; it operates based on commands and inputs.
—Efficiency: Skipping unnecessary words can make interactions quicker and more straightforward.
Ultimately, it's about what makes you feel most comfortable and how you prefer to structure your interactions. If saying "please" feels right to you, then it's perfectly fine to do so. If you prefer a more straightforward approach, that's also completely acceptable.
Well, there you have it.