Is ChatGPT intelligent?
Is ChatGPT intelligent? Let’s examine this question.
To start off, I can tell you is this: Open-AI ChatGPT has done more than just putting me out of a job. ChatGPT has recalled my vocation, eliminated my profession, and removed my very raison d’etre. I’m confronting an existential crisis.
I went to a seminary to learn how to preach. I went to graduate school to learn how to teach. I didn’t go to English language school to learn how to spell. But I can use Spell-check. I’ve thought myself to be wise and clever. But now, woe is me! The prophet Isaiah warned me.
“Woe to those who are wise in their own eyes and clever in their own sight!” (Isaiah 5:21).
Who is wiser than me? Who is cleverer? The answer is as new as it is disheartening: ChatGPT.
So, I must ask: is ChatGPT intelligent? If so, how intelligent?
Is ChatGPT intelligent enough to put preachers out of a job?
Patheos columnist Greg Smith asks whether ChatGPT can understand mysticism? Another Patheos columnist, Jim Erwin, appears enamored with Open AI in his post, “Using ChatGPT in Sermon Preparation.” Adrian Warnock went whole hog, asking ChatGPT, “Write me a sermon in the style of a Pentecostal preacher on the subject of the Holy Spirit.” As you can imagine, the resulting homily included, “praise the Lord!” The era of robot theology and digital theology is upon us. Be sure to check out AI and Faith.
With all this in mind, I asked ChatGPT to write me a sermon on grace. I’m an expert on that topic. When the sermon came back, I was stunned, astonished, flabbergasted. I’ll attach it for your reading. The sermon hit the target like William Tell hit the apple with his arrow.
No longer do I need to agonize over my sermon preparation! No longer do I need to sweat in prayer, asking the Holy Spirit to give me “utterance.” No longer do I need to rely on my theological education. ChatGPT can replace me in the church’s pulpit.
Oh wait! I’ll be out of a job. Woe is me!
Is ChatGPT intelligent enough to put professors out of a job?
Is ChatGPT intelligent enough to put professors out of a job too? I’m both a pastor and a professor. How worried should I be?
Mark Graves, a hybrid computer scientist and theologian, just might answer affirmatively in a forthcoming article to appear in Theology and Science. According to Dr. Graves, graduate students will no longer need me.
“With a fine-tuned language model for a theological area, graduate students can ask detailed questions comparing thoughts across a number of scholarly positions and improve written and oral communication through iterative learning from immediate feedback. While students would still need to formulate their own arguments, the efficiency improvements from AI-enhanced research seem comparable to the shift in physical science projects from slide rules to calculators to high-performance computers, e.g., supercomputers”(Graves, 2023).
First, I lost my job as a preacher. Now, second, I’ve lost my job as a professor. My only hope would be to assert that AI is less intelligent than I am. But am I intelligent enough to back up such a claim with facts?
Is ChatGPT intelligent enough to put scientists out of a job?
Scientists and other academics have gone into a huddle, arguing over what the next play should be. In a recent issue of the journal, Science, Jeffrey Brainard, scours the publishing playbook in vain for the winning dipsy-doodle.
Customary base defense is entirely inadequate to protect published articles or books from factual inaccuracy let alone plagiarism. Editors of scholarly journals are juggling their linebackers, not knowing for certain how to defend intellectual integrity. Should honest scientists co-author their papers with ChatGPT? Such a policy, editors fear, “could morph into a spike of manuscripts containing substantial chunks of AI-written text” (Brainard, 24/2/2023, p. 740).
Editors confront two threats. First, inaccuracy. Robot-generated text cannot distinguish between fact and ruse, error, or deceit. Might 80% of AI-generated content be trustworthy? Is that sufficient? No. Not by a long shot.
Who will do the fact-checking? Taylor and Francis publishing group is considering turning that responsibility over to the journal editors. Yeow! I’m the co-editor of Theology and Science, published by Taylor and Francis. I could not imagine taking on the fact-checking task as being proposed. Yet, how can we defend intellectual integrity here?
The second threat, as mentioned, is plagiarism. Can we detect AI-written text when it’s been edited by a cheat and integrated into a purported author’s manuscript?
To the rescue comes DetectGPT. Still under development, DetectGPT attempts to distinguish human written text from AI-generated text. Is this all an editor would need for defense?
I put DetectGPT to the test. DetectGPT analyzed that sermon below in Note #2. Conclusion: the sermon was AI-generated. It was not written either by me or by the Holy Spirit. This sermon failed the test.
“Test the spirits” says the New Testament (1 John 4:1). Should I be grateful that AI can help test the spirits for us? Whew!
Just to be fair, I also asked DetectGPT to test some of the text above in this Patheos post. Here’s the message: “DetectGPT predicts that your text is unlikely to be from GPT2.” I passed! Whew!
Now, does this mean that I’m more intelligent than ChatGPT? Or, does it mean I’m so stupid that DetectGPT could easily discern that I’m not up to AI-generated standards? So, I still don’t know the answer to my question: is ChatGPT intelligent?
Popular Harvard sociobiologist Steven Pinker seems double minded. On the one hand, ChatGPT gets confused about facts. “It utters confident confabulations, such as that the U.S. has had four female presidents, including Luci Baines Johnson, 1973-77.” On the other hand, it exhibits intelligence that nears, but falls short of, human intelligence. “We’re dealing with an alien intelligence that’s capable of astonishing feats, but not in the manner of the human mind.”
Mmmmm? I think I need to ask for opinions from more smart people. Perhaps I should go to the New York Times to find those smart people.
Ask the New York Times: Is ChatGPT intelligent?
Is ChatGPT intelligent? Some smart people writing in the New York Times, tell us that ChaptGPT is not intelligent. So, we humans need not worry that we’ll have to go extinct. Does this comfort you?
I love it when intelligent people tell me about intelligence. In this case, we’ve got Noam Chomsky, renowned emeritus professor of linguistics and cognitive science at M.I.T., whom some call the “father of modern linguistics.” Co-authoring with Chomsky is Ian Roberts, Professor of Linguistics at Downing College, Cambridge. And the third co-author is Jeffrey Watumull, Director of Artificial Intelligence at Oceanit in Honolulu, and principal investigator on programs for DARPA. Now, we have three intelligent people telling us about intelligence, both human and the artificial kind. What will they say?
If I understand these smart people correctly, they answer “no” to our question: is ChatGRT intelligent? They contend that, no matter how mind-blowing we find ChatGRT, AI is not intelligent in the sense that human persons are.
“Intelligence consists not only of creative conjectures but also of creative criticism. Human-style thought is based on possible explanations and error correction, a process that gradually limits what possibilities can be rationally considered.”
Now, did I understand them correctly? Because those three smart people are more intelligent than I am, I’m not sure. They go on to tell us that intelligent people add moral thinking to their rational thinking.
“True intelligence is also capable of moral thinking. This means constraining the otherwise limitless creativity of our minds with a set of ethical principles that determines what ought and ought not to be (and of course subjecting those principles themselves to creative criticism). … In the absence of a capacity to reason from moral principles, ChatGPT was crudely restricted by its programmers from contributing anything novel to controversial — that is, important — discussions. It sacrificed creativity for a kind of amorality.”
Is this really true? I’m not smart enough to know for sure. But they comfort me when drawing their conclusion.
“In short, ChatGPT and its brethren are constitutionally unable to balance creativity with constraint. They either overgenerate (producing both truths and falsehoods, endorsing ethical and unethical decisions alike) or undergenerate (exhibiting noncommitment to any decisions and indifference to consequences). Given the amorality, faux science and linguistic incompetence of these systems, we can only laugh or cry at their popularity.”
Is ChatGPT intelligent? Chomsky, Roberts, and Watumull answer negatively. No. ChatGPT is not intelligent in the same sense that we humans are. That’s comforting to me.
But I still have my doubts. Why? Because I’m not intelligent enough to evaluate their answer to the question: is ChatGPT intelligent?
Woe is me! What shall I do? Oh! I know. I’ll ask ChatGPT to write me an answer.
Ted Peters directs traffic at the intersection of science, religion, and ethics. For Patheos, he posts articles and notices in the field of Public Theology. Peters is an emeritus professor at the Graduate Theological Union, where he co-edits the journal, Theology and Science, on behalf of the Center for Theology and the Natural Sciences, in Berkeley, California, USA. He authored Playing God? Genetic Determinism and Human Freedom? (Routledge, 2nd ed., 2002) as well as Science, Theology, and Ethics (Ashgate 2003). He is editor of AI and IA: Utopia or Extinction? (ATF 2019). Along with Arvin Gouw and Brian Patrick Green, he co-edited the new book, Religious Transhumanism and Its Critics hot off the press (Roman and Littlefield/Lexington, 2022). He has just published The Voice of Christian Public Theology (ATF 2022). See his website: TedsTimelyTake.com.
His fictional spy thriller, Cyrus Twelve, follows the twists and turns of a transhumanist plot.
 Some years ago I wrote a scholarly paper on intelligence, “Where there’s life, there’s intelligence.” When engaged in the research, I noticed that routinely scientists tried to distinguish between higher and lower intelligence. They did not bother to define intelligence. So, I decided to provide seven characteristics that distinguish between intelligence and non-intelligence. I’ll not bore you with that here. But I will offer one brief takeaway point. What we experience as biological intelligence—especially humans but not limited to humans—includes self-agency. Is this trait found in AI or not?
 ChatGPT sermon on grace.
My beloved brothers and sisters in Christ,
Grace versus works is a topic that has been debated throughout the history of Christianity. Some people argue that salvation is a gift of grace that is freely given to us by God, while others believe that our good works are essential to our salvation. In this sermon, I want to explore this topic and help us understand the relationship between grace and works.
Let us start with the concept of grace. The Apostle Paul, in his letter to the Ephesians, tells us that “For by grace you have been saved through faith, and that not of yourselves; it is the gift of God, not of works, lest anyone should boast” (Ephesians 2:8-9). This means that our salvation is not earned through our good works but is a gift of God’s grace that we receive through faith in Jesus Christ. We cannot save ourselves through our own efforts, no matter how good or righteous we may think we are.
Grace is a fundamental concept in Christianity, and it is what sets us apart from other religions. Our salvation is not based on our works, but on God’s love and mercy towards us. Grace is not something that we deserve, but something that is freely given to us by God.
But what about works? Does that mean that our good works are not important? The Apostle James, in his letter, says that “faith by itself, if it does not have works, is dead” (James 2:17). This means that our faith should be demonstrated by our actions, by the good works that we do. Our good works do not earn us salvation, but they are a natural outcome of our faith.
We are called to be the hands and feet of Jesus in this world, to do good works and to love our neighbors as ourselves. Our good works are not done to earn our salvation, but rather as a response to the grace that we have received from God. We are saved by grace, but we are called to live out that grace in our daily lives.
The relationship between grace and works can be compared to a tree and its fruit. The tree is the gift of grace that we receive from God, and the fruit is the good works that we do in response to that grace. Just as a tree cannot produce fruit without being rooted in good soil and receiving the necessary nutrients, we cannot produce good works without being rooted in God’s grace and love. Our good works are a natural outcome of our faith, but they are not the source of our salvation.
However, it is important to remember that our good works are not the only way to live out our faith. Our faith is also demonstrated through our love, our worship, and our obedience to God’s commands. Our good works should not be done out of obligation or to impress others, but rather out of a genuine desire to serve God and to love our neighbors.
In conclusion, the debate between grace versus works is one that has been ongoing in the Christian community for centuries. However, we must remember that our salvation is a gift of God’s grace that we receive through faith in Jesus Christ. Our good works are a natural response to that grace, but they do not earn us our salvation. We are called to live out our faith in our daily lives, to be the hands and feet of Jesus in this world, and to love our neighbors as ourselves. Let us remember that we are saved by grace, and let us live our lives in response to that grace with faith, love, and good works.
May God bless us and guide us in our journey of faith.
Amen. My co-editor is Robert John Russell. Bob along with Braden Molhoek at the Center for Theology and the Natural Sciences (CTNS) have just launched a three year research project, “Virtuous AI: Cultural Evolution, Artificial Intelligence, and Virtue” (Russell, 2023).
Brainard, J. (24/2/2023). Journals take up arms against AI-written text. Science 379:6634, 740-741.
Graves, M. (2023). ChatGPT’s Significance for Theology. Theology and Science 21:2.
Peters, T. (2017). Where There’s Life There’s Intelligence. In e. Andreas Losch, What is Life? On Earth and Beyond (pp. 236-259). Cambridge: Cambridge University Press.
Russell, R. J. (2023). Artificial Intelligence: What can we learn about being human from non-human technological ‘life’? Theology and Science 21:2.