By now, you have probably seen reporting on A.I. (Artificial Intelligence). The format of the reports are almost humorously consistent. There are a few moments of explanation on how computing reached this point. There is a mention of A.I. being the quickest technology to reach one hundred million users. Then, the report often concludes with the pundit explaining that a portion or all of the report was written by A.I. So, before I proceed. No, this blog was not written by A.I.
Viewing these developments with some level of skepticism, I thought I would test A.I. out myself. I had worked as a UNIX architect for almost a decade before responding to my call into ordained ministry. I still enjoy using technology to accomplish tasks. Recently, I attended a Board of Ordained Ministry meeting where a few of my colleagues posited that A.I. performed better on some of the commissioning questions than several of the candidates. Could this be right? So in preparation for a Sunday School Assembly that next week I was giving on United Methodism's future, I tried A.I. out. I do not know what I was expecting, but I was surprised by the cogent and clear answers it provided along with a useful summary recap.
It was a 'Hello world!' moment. For those who do not know, the output 'Hello world!' is often the first step in learning to code. For those without this background, I would liken the feeling to the first time you saw a computer or held an iPhone. I was not sure what, but something had forever changed.
Summaries are one thing, but how would A.I. do in a deeper set of questions. For the next sermon I was preaching on Matthew 10:26-39 some of the difficult passages of Jesus, I asked the A.I. to provide me three questions on the text. All three questions were questions of understanding.
- What is the meaning of the phrase "fear him who can destroy both soul and body in hell" in Matthew 10:28?
- In Matthew 10:34, Jesus says, "Do not think that I have come to bring peace to the earth. I have not come to bring peace, but a sword." What does he mean by this?
- According to Matthew 10:39, "Whoever finds their life will lose it, and whoever loses their life for my sake will find it." What is the significance of this statement in the context of the passage?
You will notice that each of the questions ask questions of meanings or significance. And, this seems consistent with what most people are seeking when they encounter the text on a Sunday morning.
I then asked Chat GTP to take those three questions and modify them to build someone's faith. Wow! It asked some surprisingly good questions.
- How can we trust in God's protection, even when facing difficult and frightening circumstances, as emphasized in Matthew 10:26-28?
- In what ways can we embrace the transformative power of the Gospel, even when it causes division and conflict within our relationships, as described in Matthew 10:34-36?
- How can we surrender our lives to Jesus and follow Him wholeheartedly, even when it requires us to let go of our own desires and comforts, as encouraged in Matthew 10:39?
Now I was impressed. I asked it to give me a five minute sermon. It did. And, it was a bit direct and not particularly creative. But, the sermon generated did address the questions raised above. I then asked Chat GTP to add modern theological references and it added quotes from Dietrich Bonhoeffer and Timothy Keller. The quotes were insightful and appropriate. It was then that what I believe will be the real power of A.I. emerged. Those quotes prompted me to remember several other quotes, one of which I utilized in the message. The questions that were asked helped me frame the topic.
Although the sermon itself was not particularly useful, the questions and insights were. Why would this be so? A.I. is designed to provide insights from very large data sets. The algorithms of Chat GTP's A.I. are aggregations of all open data on the internet. Imagine that you could have access to the thoughts of billions of people living and dead. Those aggregate thoughts would then be sifted through and simplified into a few questions that any reader with basic comprehension could understand and be clearly presented to you. Would that be helpful to your sermon preparation? I think the answer is yes.
But with the clarity and hidden preferences for what data is being sampled, there are also a few potential challenges. The output a learning algorithm can produce from aggregate data is only as good as the input. For example, if the data of the internet is skewed by the social setting of those producing it, the hidden biases will be be present in the output. Issues of race, class, gender and how the Bible calls us in our prophetic voice can be diminished if we are not making that an intentional part of the questions we ask in our reflective preparation for preaching. Adding questions from the A.I. can be helpful as long as it does not simply reinforce existing biases and blind spots. A.I. also tends to mirror the loudest or dominant voices in society as captured on the internet currently.
But, even reflecting through this theologically helps create a better sermon. What I found helpful was the creative thoughts that these questions caused me to reflect on. I reflected on the surface questions of understanding more than in previous messages. When preaching for many years it is easy to forget how much you know about a text that you assume is known. I was also caused to ask myself how the message I was preparing addressed the more existential questions that the A.I. posed when I asked for questions that could build faith. And, I found myself considering how non-dominant voices and questions would challenge the A.I.'s questions and thoughts on the text.
A.I. will not replace pastors in the task of preaching. But, A.I. can assist pastors in their process of reflection and preparation. My task this week will be to ask the A.I. the questions from my introduction to preaching class and see how it does. I will keep you posted. At the least, I expect to be using A.I. in my preparations. And, pastors who do not will likely be missing something.
Comments
Post a Comment