Recently the AI tool ChatGPT has exploded onto the scene, ushering in a first generation of free and useful artificial intelligence tools online. AI has been a fixture of science fiction books and films for nearly a century. Perhaps you’ve seen the classic 2001: A Space Odyssey, where the HAL9000 computer has realistic conversations with astronauts. Or more recently The Matrix suggested that the world we live in is an elaborate computer simulation. What is the state of AI? Can we really create something that can reason like a person? Does the Bible give us anything to draw from on this topic as faith-centered leaders?
What is AI? Artificial Intelligence (AI) refers to application or computer system that thinks, reasons, and learns in the same way as a human being. The key characteristics of AI include a computer system that solves problems in much the same way that people do, and can self-develop—that is, increase its knowledge and ability through experience so that it does better the next time. And like the human brain, AI has instant access to lots of information.
In computer and app development, we have reached a point where CPUs are fast enough, mass storage is large enough, and programming sophisticated enough, that we can mimic facets of the human brain. Medical research conducted by Stanford University has quantified the memory capacity of a typical human brain as the equivalent of 2.5 petabytes of data (or 2.5 million gigabytes). The cerebral cortex contains 125 trillion processing synapses.[1] These capacities, while incredibly large, are now becoming reproducible in computer chips and storage devices. The most advanced computer chip today contains 2.6 trillion transistors and 850,000 processing cores[2]—approaching the range of the human brain in broad scale.
What is the limitation of AI? It’s important to understand that the very best AI software and hardware today only attempts to simulate the human brain in size, scale, and processing capacity. In some ways AI can outperform the brain. For instance, it can consider every possible solution to a problem systematically over a number of minutes, hours or days. And it can access an array of data from which to draw a solution from the entire connected Internet, which in 2023 stands at roughly 5 million gigabytes of information.[3] It’s like having every available book, encyclopedia and reference ever written at the ready. A computer’s ability to run a simulation or process a complex computation are unparalleled. The human brain cannot do it.
In the 1960s, author Arthur C. Clark imagined a computer so advanced that it took on human characteristics. In his story the HAL9000 computer eventually went homocidal and killed several astronauts.
What the brain can do, however, is make creative and original connections between points of data. An experience in childhood, and a problem at work. A feeling you got from your spouse, and a thought while waiting for a meeting to start. The brain’s ability to understand nuance and emotion are not replicated in any meaningful way in even the most advance AI. Consider too how a human understands the world.
Our intelligence relies on the way we view the world as people, and the way we think about the world. AI has not yet dipped its code into areas like common sense or emotional intelligence. AI can perform tasks and recognize patterns. But the human experience is such an enigma, that when you ask even the most advanced AI a question, it has no means to interpret how you may feel when you ask it. Also, scientists and programmers don’t yet understand facets of the human brain, much less how to replicate them. For instance, scientists don’t yet agree on why we sleep, or why dreams manifest unconsciously, or how consciousness begins and ends, or where savant or virtuoso talents come from.
Some computers appear smart, but are really just relentless in their processing. In 1997 IBMs supercomputer Deep Blue defeated the World Champion of the time, Garry Kasparov, in a chess match under tournament conditions. It wasn’t that the program was thinking creatively, though. The software simply had so much computing power that with each move, it could consider every possible outcome—billions of potential moves in a moment—and then choose the best option. Kasparov could consider less than a hundred possible moves in the same time.[4]
Since the 1950s, scientists have used a standard to judge whether a computer is truly intelligent: The Turing Test. Developed by Alan Turing, the test is simple: In a blind setting, you ask a human to tell the difference between an artificially intelligent interaction and another human. If someone can’t tell the difference, the AI passes.[5] While ChatGPT is incredibly useful and displays human characteristics in its responses, after working with it for a bit you begin to see the cracks. It’s somewhat formal in its answers. Its writing contains idioms and patterns that are designed to appear fluid and natural. But it speaks in a way that no person really would. And most importantly, it has but one direction or opinion on a topic. Ask a person a question and they can give you an entirely different answer on different days. ChatGPT, however, is overly consistent. It thinks one way. So a person can indeed recognize it as artificial.
How should a Christian approach AI? Will AI tools continue to develop, and become more natural and useful? Of course. Will we eventually make the quantum leap forecast in the Terminator movies, where machines become self-aware and start making decisions for themselves? Scientists agree this is a long way off, if not impossible, given our limited understanding of the mind itself. Quantum computing, which is a massive leap in computing power, is in its infancy, and many years from being practical, and many years beyond that to realize quantum applications.
Artificial intelligence has yet to develop a sense of purpose—to have a meaning for being that was not programmed into its algorithms. AI cannot contemplate its own existence, or consider how it was created. If AI came to understand in some rudimentary way that it was made by people, could it then comprehend Who made the people who created it, or the source all of the matter from which its various hardware components were made? AI is not capable of holding beliefs, a personal set of values that gives order and direction to its existence. Finally AI has no sense of personality, the uniqueness of each human being to be considered an individual with hopes, dreams, personal style and preference—to say nothing of how a machine would understand feelings. In movies about AI, the deepest emotion expressed is fight-or-flight, simple survival. Indeed this response is not even emotional in a person, but a function of the autonomous nervous system. What could love or empathy ever mean to a machine intelligence?
A recurring theme in AI movies… In the Terminator series the robots become self-aware and decide to launch a nuclear holocaust. AI is often scary because we assume its first emotional response, if it could have one, would be flight-or-flight, simple survival. And humans would be the threat.
A Christian should consider AI to simply be an advanced computing tool. Elements of this technology will prove useful for life and work, and even ministry. Many repetitive jobs that don’t take creative thinking could be replaced by AI programs. Fast food franchises are experimenting with restaurants that contain no employees. Intelligent programs talk to you at the drive through, robots process and cook your order. This is not a tremendous leap for current technology. And who really wants a career in fast food anyway?[6] Food, farming, cleaning, restocking and retail—there are many areas where AI may ease the burden of work for many and create new opportunities for people to use their talents, gifts and abilities in a more meaningful way.
An interesting Scripture that perhaps applies here is Romans 1:21-23: “For although they knew God, they did not honor him as God or give thanks to him, but they became futile in their thinking, and their foolish hearts were darkened. Claiming to be wise, they became fools, and exchanged the glory of the immortal God for images resembling mortal man and birds and animals and creeping things.” In other words, man, instead of worshipping a holy God, creates things resembling himself so that he can pay homage to his own creation. Man wants to express himself as God so he does not have to bow to One. If we can just make something that appears utterly close to human, then we can stop acknowledging God is Who He says He is. Look at how brilliant we have become—surely there is no immortal God over us!
And this is perhaps what we will see from AI… a desire on the part of the people working on it for it to be more than it really is. A sophisticated mix of hardware and software that has useful applications and more natural interactions? Yes it is. A real intelligence that exists with its own purpose, meaning, beliefs, emotions and dreams, that learns and grows over time, gaining in wisdom and experience? When we think about the miracle that is life, and the complexity that is man, and the greatness of God, our most advanced dreams of AI fade. But, at least now we have ChatGPT to discuss an interesting concept!
[1] https://www.medanta.org/patient-education-blog/what-is-the-memory-capacity-of-a-human-brain/
[2] https://www.tweaktown.com/news/74601/the-worlds-largest-chip-2-6-trillion-transistors-and-850-000-cores/
[3] https://www.easytechjunkie.com/how-big-is-the-internet.htm
[4] https://www.ichess.net/blog/best-chess-engines/
[5] https://tech.co/news/artificial-intelligence-human-brain-2018-04
[6] https://www.franchisewire.com/automated-mcdonalds-in-texas-generates-mixed-reviews/