Artificial Intelligence And Our Future School

I. Honesty

Our faculty online discussion group exploded the other day when some of us read the recent article "The College Essay Is Dead" from The Atlantic magazine about the ChatGPT Bot [1]. Chat Bot is an Artificial Intelligence (AI) program that I was able to sign into on my computer in about three minutes. You can ask it any question and it will give you a perfect, three paragraph essay if you want. The New York Times touted, “The latest natural-language system generates tweets, pens poetry, summarizes emails, answers trivia questions, translates languages and even [ italics added] writes its own computer programs.” No self-respecting ChatGPT Bot would abide by their adverb use there.

There’s an elephant in that room. Its trunk is shaped into a question mark: will students use ChatGPT Bot to write their college essays? Will students use it to write their research paper? Is real writing dead!? Is thought obsolete! Will plagiary go viral! Why bother!

Our Environmental Science teacher, Nick Scacco, was concerned that “These free AI chat bots can generate in 10 seconds what it would take me hours to generate.” Nick’s too nice to say it this way, but we’re all concerned about plagiary and student motivation to produce original work. Our Government teacher, Jessi Brown, posits: “I guess we're going back to all handwritten assignments?”

To challenge this threat, I dove straight into it. First, I asked Nick to compare my last column to the ChatGPT Bot version. In the future, could I just ask it to write my column? Nick quickly posted the AI version of my column on coping with threats of violence in America, and my fears of AI rendering our brain obsolete were quickly dispelled. ChatGPT Bot: 

• Cut my verbiage in half
• Cut out personal reflection about my mentor, Dr. Edith Eger
• Cut out my alternative/creative thinking and unique phrases I made up
• Cut out any reference to guns, which is a huge issue in America
• Abandoned anything like writing style

In general, ChatGPT Bot turns a reflective story into a logical, sequential list, perfect for the SATs and for machine readers, with no real rhythm or pathos. It spit out an automatic A paper with no charm or original thought, no risk, and no rhythm. Completely predictable. All the medicine of a placebo. The Chat Bot coverage of my column was more of a “how-to” book entry than a think piece. The Chat Bot served up the very reason I am always fighting with our English department to abandon the requirement for the standard 3-paragraph essay with its three perfect topic sentences.

I'm not saying my writing is better. ['Tis.] I'm only saying I got no joy reading the Chat Bot version… except maybe in the ironic thought that only a machine essay reader (which is being used nationwide) would be reading it. 

I took aim. Here are three Chat Bot prompts I made, with the following Chat Bot answers. English students could write stories about these answers, Science students could research them, Philosophy and Psychology students could burrow into them:

Prompt: Why is ChatGPT terrible for education?

Chat Bot response:

In general, language models like ChatGPT are not well suited for education because they lack the ability to provide accurate and reliable information or to engage in meaningful interactions with learners. They are simply programmed to generate text based on the input they receive, and they do not have the ability to provide in-depth knowledge or expertise on a given topic.

Prompt: Is ChatGBT a form of thought?

Chat Bot response:

I am not capable of having thoughts or feelings in the same way that humans do. I am simply a tool that can be used to generate text based on the input that I receive. I can provide information and answer questions to the best of my ability based on my training data, but I am not capable of experiencing or expressing emotions or engaging in conscious thought.

Prompt: What is thought?

Chat Bot response:

Thought is the process of using your mind to consider something, typically involving conscious mental activities such as analyzing, reasoning, and conceptualizing. It is a fundamental aspect of human cognition and is essential for learning, problem-solving, and decision-making. Thought can be both conscious and unconscious, and can involve a wide range of mental activities, including perception, attention, memory, language, imagination, and emotion.

Conclusion: In the future, Chat Bot and artificial intelligence may encourage teachers to focus students far more on their personal thoughts—thoughts that Chat Bot can’t have. Students ought to be writing (and sure, Jessi, even handwriting) to access their sensations, aliveness, and all things Chat Bot would remove from the essay. 

II. Truth

Something else it will take the human mind to determine that a Bot can’t: truth. Autocratic-oriented politicians such as in the Philippines (Marcos and Duterte), Russia (Putin) and the US (Trump) are using vast social media outreaches to claim lies are true, and we have no way of preventing Chat Bots from spitting those lies up in answering our requests for information. ChatGPT Bot doesn’t actually say where it found a piece of information.

The point is: information is incredibly easy to get these days. The valuable commodity is not information, it is trust. Students don’t have to create information, they’re swimming in it, but they do have to find real ways of being gatekeepers of truth, honest, and civility. That makes their job harder, not easier. Let’s think about the acronym: AI. What’s that first letter there stand for? Artificiality is exactly what we don’t trust.

Teachers and students have spell check. We have plagiary check. So how about fact and lie check? That's the Bot we need!

An MIT study in 2018 showed that lies spread six times faster than facts! [3] (Facts are less inflammatory so we don’t tend to pass them along.) With ChatGPT Bot around, that multiplying of lies could increase. Teachers and college admissions officers will have to ask better questions that challenge students to write what is true. Dealing with that risk could be exactly the challenge we need.

Determining if a piece of information is actually correct or not is already a monumental task, hence the mere teacher prompt: “Where did you get your information?” has clearly become “the” ultimate 21st century teacher question, and ChatGPT Bot makes the question more essential still. 

III. Beyond AI

Perhaps more inspiring, teachers will have to go deeper into the nature of thought and consciousness with students willing to explore that realm—we have ChatGPT Bot to thank for that, too. For their parts, students will need to go deeper into themselves: “Know Thyself,” the ancient precept commands. This is not something a ChatBot can do, and yet there is nothing more valuable a great teacher can ever teach.

The human brain is vastly more complex than any AI reproductions like ChatGPT Bot, but the computer tends to define, hence limit, what we think about. If we start at the edge of that limit, AI could help us launch ourselves into a world of wild creativity, passion, and courageous imagination that takes us beyond the “Bot”. Happiness and peace of mind have everything to do with balancing conscious and subconscious thought and Chat Bot, of course, has no subconscious. 

True, Chat Bot could replace Wikipedia. Chat Bot is not original thought, although Bots can combine ideas in new ways. Nor is ChatBot necessarily true and referenced thought, though it is a baseline of widely accepted information from which we can take off with students. “Start with a Chat Bot answer, and make it real, alive, and personal,” might make a great prompt for any class—“Flush out the Bot answer with personal story.” The Chat Bot is the prompt, the launch, the information we are looking for. Chat Bot is not the answer our students need, but it can be the question.

[1] "The College Essay Is Dead", The Atlantic, Stephen Marche, December 6, 2022.

[2] "Meet GPT-3. It Has Learned to Code (and Blog and Argue).", The New York Times, November 24, 2020.

[3] "Study: False news spreads faster than the truth", MIT Management, March 8, 2018.

COMMENT! Click on the "Comments" drop-down box below to share a comment about this column.