“The naïve believes everything, but the sensible man considers his steps.” – Proverbs 14:15
When writing came along to the ancient world, Plato lamented that people would no longer commit things to memory or engage with each other. Chinese philosophers warned against the danger of a new technology called “books.” With each new invention, there are growing pains while society learns to adjust and humanity works out how to glorify God with the new technology. Today ChatGPT and other large language model AI’s have quickly become an overnight sensation. Those that promote and defend this kind of AI say that it is “just a tool” that can be used for good or for ill. The problem lies in understanding just what kind of tool it is. To be clear, this article is not condemning AI as a technology and it can surely become more useful as it develops. For now, here are five reasons to be prudent about generative AI.
- AI is generic and repetitive—The answers that AI give are impressive and seem detailed, but a careful examination shows the written responses to be lackluster. While a non-writer may not be able to see the difference, it is always stronger when a person attempts to put things in their own words.
- AI is untested—This type of AI is still in its infancy. There simply has not been enough time or use cases to say that the information provided is accurate or helpful in the long term. The technology is bound to improve, but consider the current iterations to be like rough drafts.
- AI is “of the world”—Large language models scour the entire internet to return its answers. Anything related to faith and biblical accuracy should be treated with the utmost suspicion until verified by primary and trusted sources. The bias of the world is bound to creep in given the way AI finds its answers.
- AI is personally impersonal—AI is design to mimic conversational tones between two people, so it is easy to forget that what is giving answers is a program, not a person. This ability to mimic is imperfect, however, and can’t replace the individuality that a human brings when creating content. The net result is, like point #1, a generic lackluster set of answers that has no personality.
- AI is tempting—When one considers the time savings of letting generative AI do the “heavy lifting,” it is a short step to actually using it for unethical purposes. Examples include passing off writing as one’s own that is not and accepting answers as sacrosanct without truly knowing their sources.
Combined, these reasons illustrate the danger for those of us in ministry using AI to create content and resources that are mediocre, impersonal, worldly, and factually inaccurate. This kind of AI is a tool, but some people are eager to use it incorrectly—like using a hammer to craft a detailed sculpture. It may be helpful for the most simple of tasks, but when it comes to the work of God’s church and reaching others, we must make every effort to perform at our best and conform to God’s will. And in our weakness, we should turn to the Lord, not a large language model.
This article was written 100% by a DiscipleLand human and is solely the opinion of that human.
