Every Careless Word We Generate?

Jesus tells us that, on the judgment day, we will have to give account for every careless word that we speak (Matt. 12:36). I suspect that the words we write will be judged with equal or greater severity, for written words are often more permanent and intentional. If Jesus’s warning against careless words extends to our writing, it is logical to assume that it also extends to our use of Generative AI and Large Language Models. It would thus do us well to carefully filter the words we speak, write, and generate.

Lazy Language Models

Notice that Jesus does not simply say every untrue or harsh word will be judged but every careless word. This is a frighteningly broad term, warning us against the carelessness of insults, lies, gossip, profanity, and sarcasm. Our speech and writing must be good, true, and beautiful. Accordingly, should we choose to use AI tools, we must do so with the intention of communicating what is truthful, useful, and kind.

But the word for “careless” in Matthew 12:36 can also be translated as “lazy.” This is the biggest issue I have with Generative AI and LLMs; they present endless opportunities for lazy words. What is more lazy than using AI to write articles or blog posts when we ourselves could do so, albeit at the cost of our time and effort? Or having Gemini write emails to our employees instead of learning to communicate with them ourselves? Or having ChatGPT outline a term paper when we could have done so with just a little more forethought and time management?

I’m reminded of an advertisement that ran during the 2024 Summer Olympics. It’s only a minute and I encourage you to watch it below.

This advertisement begins with a father describing his young daughter, who is a runner and looks up to Sydney McLaughlin-Levrone. But this heartwarming tribute takes a discouraging turn when the father admits that, although he’s good with words, he would prefer for Google Gemini to help his daughter write a letter to her hero. He proceeds to dictate the type of letter he’d like his daughter to write, and Gemini produces a perfect—if impersonal—epistle.

Isn’t it strange that an advertisement broadcast during the Olympics—which celebrate hard work—would so prominently feature carelessness? This dad didn’t give his daughter the chance to write her own letter, or even to try. Gemini didn’t help her learn to write a letter; it did it for her. The daughter didn’t even type in her requests to Gemini; her dad did.

This advertisement is the epitome of carelessness. We have to hope that the daughter puts more effort into her running than her writing, but Scripture reminds us that if we are not faithful in the little things, we are unlikely to be faithful in the big things (Luke 16:10; 19:11–27). If we cannot wrestle our way through a brief letter, we are unlikely to win our way to the Olympics.

A Potential Positive

On a more positive note, AI can be used to combat careless writing. Do you ever know that there’s a word for a concept you’re trying to communicate, but you just can’t think of it? A major benefit of LLMs is their ability to conjure up precise and accurate words from our jumbled definitions. In this sense, they serve as super-powered dictionaries and thesauruses—and I don’t know anybody who is morally opposed to ye olde reference books. (For example, I just had to look up the plural of “thesaurus” because I was unsure if it was “thesauri” or “thesauruses.” Google quickly assured me that either is fine.)

If we can use AI purposefully, it can be a great tool for finding “good, strong words that mean something”—to borrow a phrase from Anne of Green Gables. Using AI to clarify our syntax, spelling, and vocabulary seems appropriate because such uses evidence a desire to write with care.

Conclusion

As AI becomes increasingly accessible and accepted, I urge you not to generate careless words. Write and, if need be, wrestle your way through your emails, social media posts, and school papers. I doubt that Jesus will mind if your final draft is not as good as Google Gemini’s would have been, but he is deeply concerned with the care that you—a human being—put into your words.

If you can use AI to refine your writing by checking for spelling or grammatical errors, by all means do so. But take heed lest you use AI to avoid rather than refine your work.

Thank you for reading! If you enjoy my writing, consider subscribing for free or sharing this post with a friend.