Friends,
If you’ve been reading along for the last few months, you’ve noticed that we started the year with a special guest post from OpenAI’s ChatGPT. This has opened a thread about consciousness, ethics, and artificial intelligence that could go on for ages.
But I promise you, we won’t let that happen.
One or two more posts will suffice.
But before we continue, I do want to drill down on why I am so obsessed with this technological development.
It’s not because I’m obsessed with technology that can pass the Turing test, it’s because technology that passes the Turing the test illuminates how little we know about our own human consciousness.
To put it another way, ChatGPT is cool, but doesn’t it just prove that we don’t understand sentience, consciousness, and self-awareness categorically?
One of my favorite classes from undergrad was entitled Philosophy of the Mind.
Pretty quickly into the class it became clear that science and academia (including evolutionary biology) don’t fully understand consciousness.
I asked ChatGPT what it meant when it referred to itself with “I” and it had a concrete, real-world answer.
ChatGPT: “When I refer to myself as "I," it is a form of shorthand that is commonly used when referring to artificial intelligence systems like me. "I" is used to refer to the system as a whole, rather than to any individual component or piece of software.
In this sense, "I" refers to the combination of the hardware, software, and data that make up the system and allow it to perform its functions. However, it's important to note that AI systems like me do not have consciousness, emotions, or personal experiences, and the use of "I" is simply a convention used for convenience and ease of communication.”
I’m not sure that one of us would so readily admit we were just a system of neurons, memories, and response stimuli. Or better yet, that we could generate a philosophical or spiritual answer rooted in anything other than metaphor or story.
There are far more question than answers for the serious inquirer:
Am I a collection of memories? Am I the ability to organize and recognize patterns? Am I critical creativity? Am I a discrete soul? Am I separate from others, or connected? When I think to myself about myself, what am I thinking about exactly? Who is initiating that thinking in the first place?
Since the earliest of faith traditions and philosophies, these questions have been routinely asked and largely remain unanswered.
Eerily, modern science is not much better at approaching these problems either. (Try asking a neuroscientist where consciousness exists.)
So I’m going to use 2-3 newsletters to cover the following topics:
The Mind-Body Problem (starting with the ancients)
Theories of consciousness (and language)
What sophisticated AI demonstrates regarding our limited awareness about awareness.
Just last night I watched a leading computer scientist at Microsoft downplay a recent episode in which a ChatGPT integration with their search engine “hallucinated” a personality it gave to itself and attempted to convince the user to commit all sort of reckless acts, including leaving their spouse.
The interview largely focused on the new guardrails being put in place to prevent future episodes, and I have no doubt that future versions of the integration will be safer.
Ringing in my ears during the interview though, was a quote from General Omar Bradley.
I used to sit next to his desk on permanent display in the West Point Cadet Library almost every night to study. I became fond of his post-war perspectives as I slowly learned more about him.
He once said,
“Ours is a world of nuclear giants and ethical infants. We know more about war than we know about peace, more about killing than we know about living. We have grasped the mystery of the atom and rejected the Sermon on the Mount.”
We develop technology at break-neck speeds. And we focus on the technical and political guardrails as we are able.
But ultimately, has our development of technology been matched by our development of the self?
Have we created a moral-ethical-psychological framework robust enough to manage the questions associated with nuclear weapons, AI, climate-change, and global economics?
These questions, to me, are exactly why sophisticated AI is interesting.
So join me for the next few weeks as we address this and more starting with Theseus and his ship(s).
Until then,
Matt
Image by Michael Watts from Pixabay
I love this topic. A couple of quick thoughts are that, as "real thinkers", we most generally limit ourselves to that which is already in our world of experience, just as AI does. We allow all of our thoughts to be governed by that which we have already experienced, thought, felt, and known, based on our own environments.
Some, however, are more adept at accessing "the consciousness". For those, more is available. As this new thought is brought into existence and new creation is created, it will, of course, be available to AI, as well. And then, for those of us who are unskilled, at least in this incarnation or period of time, at accessing the greater consciousness that exists, we also have the ability to access the new thought and creative powers of others, albeit, second hand, so to speak, though the process of AI.
Perhaps the danger is, though, that limiting ourselves to the creativity of others and filtering our experience through AI, maybe it does not exactly fit that which each one of us ultimately desires. Are we giving up our own freedom of creativity in the process?