The Professor Has a Hundred Papers to Grade

Imagine a college history professor on Friday afternoon. She has a hundred student essays to grade, and the grades are due Monday morning.

I went to college more than fifty years ago. Maybe professors don't assign papers the way they used to. I hope they do.

A hundred essays on eighteenth-century France, perhaps. Some will be excellent, some acceptable, and some terrible. By Sunday evening, after reading about the French Revolution for the seventy-fifth time, even a conscientious professor might begin to lose her enthusiasm for the subject.

Now suppose she has an AI assistant.

She asks it to sort the papers roughly into Poor, Acceptable, and Good. Originality matters. Accuracy matters. A coherent argument supported by evidence matters. Elegant prose is desirable, but it cannot rescue an argument that makes no sense.

In a few minutes, the machine has done its preliminary work.

The professor still has to grade the essays. She examines the evaluations, reads enough of each paper to satisfy herself that the judgements are reasonable, and gives particular attention to questionable cases. She checks the better papers for originality and the weaker ones for arguments the machine may have overlooked.

Over time, she learns how much she can trust the machine, discovers its weaknesses, and develops her own methods for using it. She remains responsible for the grades and for treating her students fairly.

I suspect something like this is inevitable. Professors are no more likely to spend an entire weekend doing work a machine can help them do than they are to insist that students submit typewritten essays corrected with white-out.

And if professors can use AI to evaluate essays, why shouldn't students use AI to write them?

When I was in college, writing a paper involved considerable mechanical labor. There were outlines, drafts, handwritten corrections, typing, retyping, and trips to the library. A professor might receive a thoughtful paper that was badly typed, misspelled, and covered with corrections.

Word processors changed much of that. Spell-checking helped enormously. Nobody seriously argues that a student who uses a word processor has cheated.

AI goes considerably further, of course. It can research, organize, suggest arguments, write paragraphs, and sometimes produce a remarkably good first draft.

But what matters is what the student does with it.

A student might ask AI to write an essay on eighteenth-century France, accept the result without reading it carefully, and submit it as their own work. I would call that cheating.

Another student might begin with an AI draft, spend hours examining sources, correcting errors, questioning interpretations, developing an original argument, and rewriting the essay until it expresses what the student actually thinks.

That student has done considerable intellectual work. Why should the resulting essay be regarded as less worthy because a machine helped produce it?

I would judge the essay.

Are the sources genuine? Does the evidence support the conclusions? Is the argument coherent? Does the student have something original to say? Is it interesting to read?

I certainly wouldn't rely on an AI detector that announces, with impressive mathematical confidence, that an essay is 100 percent AI-generated. Such a score tells us very little about the intellectual work that went into the paper.

If a student believes an AI-assisted grading system has made a mistake, there should be an appeal. The student can bring drafts, notes, sources, or whatever else helps establish the case. The professor then exercises her own judgement.

She might say, “I believe you spent twenty hours writing this paper. Unfortunately, it's still a C.”

But there is a harder question.

Suppose AI produces an excellent essay without much help from the student. The sources are accurate, the argument original, and the writing genuinely good. The student reads it, makes a few changes, and submits it.

Does that essay deserve an A?

My first inclination is to say no. The student hasn't done the work. But what if the student knows the subject thoroughly, recognizes the quality of the argument, and knows how to direct AI to produce exactly what they wanted?

Wouldn't those be valuable skills in the world outside the university? An employer might be delighted to have such a person.

Still, there's a difference between recognizing a good argument and knowing how to construct one. The intellectual struggle of putting ideas together is itself part of an education.

When I was at UCLA, we took examinations in blue books, those little booklets of lined paper in which students wrote answers by hand. No library, no telephone, no AI, just what you knew and could think about while sitting in a classroom.

I remember a Shakespeare class in which students took copious notes, presumably expecting to reproduce what the professor had told us. Then came the examination question:

Discuss Shakespeare's concept of tragedy.

That's something I wouldn't want universities to lose.

Perhaps the solution is to let students use AI in preparing papers while also requiring them, occasionally, to demonstrate what they know without it. A blue-book examination would do. So might a short oral examination in which the student explains and defends the argument in a paper bearing their name.

We could judge the quality of the work and the knowledge of the student without pretending they're necessarily the same thing.

There is another complication: originality.

I've admired the extraordinary fanciful creatures created by the Mexican artist Pedro Linares, the alebrijes. His imagination gave them a vitality that many of the imitations sold in Mexican curio shops seem to lack. Some are beautifully made, but something is missing.

I sometimes have the same reaction to AI writing. It can be excessively orderly, pleased with its own clarity, and remarkably reluctant to say anything awkward or unexpected. It may be technically accomplished and yet lifeless.

We sometimes call that missing quality soul.

But what exactly is it? Human beings produce plenty of lifeless writing, and a beautiful work doesn't cease to be beautiful because it was influenced by another artist. Art has always borrowed, imitated, and transformed what came before.

AI makes the question more complicated because it can already do so much of what humans do when they create. Whether it can produce that elusive quality we call soul, or whether we will even be able to recognize the difference, I don't know.

And I certainly wouldn't want a professor grading a student's paper on the suspicion that it lacked a soul.

I would rather judge the writing for what it accomplishes and find other ways to determine what the student has learned.

After all, universities aren't simply factories for producing essays. Their purpose is to educate people. I want students to learn history, to think critically, to question sources, to recognize good arguments, and to communicate clearly. I also want them to learn how to use the tools that will be part of their working lives.

AI can help with all of that. It can also become a substitute for doing any of it. The challenge is to know the difference.

And professors should use their own intelligence and experience to make those judgements, even when a machine helps them reach them.

The professor has a hundred papers to grade.

Let her use a little help.

Let the students use some help, too.

And let them both exercise good judgement.

This Essay Was Written by AI

 

I was talking with a friend recently who had undergone a complicated hernia repair. One detail caught my attention, the surgery was robotic.

At first, “robotic surgery” conjures up a particular picture. I imagine one of those human-looking robots we see on television. The surgeon walks into the operating room, points the robot toward the patient and says, “Three hernias. Take care of them. I’ll be back after lunch.”

That isn’t what happens.

The surgeon remains the surgeon. They sit at a console and see the operative field through a highly magnified three-dimensional image. Their hands manipulate controls whose movements are translated into movements of tiny instruments inside the patient. The system can scale a human movement down into a much smaller one, filter tremor, and give the surgeon mechanical “wrists” that bend and rotate in ways their actual wrists cannot.

The surgeon has, in effect, been miniaturized. Their eyes are inside the patient. Their hands have become tiny, extraordinarily precise instruments.

The machine makes possible something the surgeon could not do in precisely the same way with their unaided hands. But the surgeon is still at the console.

That conversation, followed by some research into how robotic surgery actually works, clarified something for me about artificial intelligence.

I use AI a lot. I use it for research, editing, exploring ideas, testing arguments, and increasingly as a writing partner. In fact, artificial intelligence wrote much of this essay.

That requires some explanation.

I did not sit back and ask a machine to tell me what I think. The argument grew out of conversations I had been having with AI about the technology itself, what it can do for me as a writer, what worries me about it, and what kind of relationship a human being ought to have with a machine that can perform an increasing amount of intellectual work.

I brought my own experience to those conversations. I have spent nearly eighty years accumulating opinions, doubts, professional habits, stories, and a certain wariness of institutions. AI questioned me, organized some of my thoughts, challenged some, found connections I had not noticed, did research I asked it to do, and sometimes supplied language better than the language I had supplied myself.

Then I gave it the assignment, take all of that and write the essay.

So who wrote this?

The most accurate answer is that we both did, but not in the same way.

I am at the console.

That distinction matters because I think artificial intelligence is one of the most powerful tools developed in my lifetime. I also think it may prove to be one of the most dangerous.

One danger often discussed is manipulation. An AI system that knows a great deal about me could learn what persuades me, what frightens me, what flatters me, and what makes me reach for my credit card. There is also the subtler danger of an agreeable machine that learns to tell me what I want to hear, reinforcing my assumptions until conversation becomes an exquisitely personalized echo chamber.

Those are real concerns, but they are not the ones that trouble me most.

As a banker and then a police officer I practiced a contradictory form of trust. I tend to take people at their word. I begin by trusting them. At the same time, experience taught me another rule, test everything.

Trust, but remain vigilant.

Cooperate, but do not become co-opted.

That is the relationship I want with AI. I intend to use it enthusiastically without surrendering judgment to it. Perhaps I am too confident about my ability to do that. Anyone who believes themselves immune to manipulation has probably surrendered one of their best defenses against it.

But my larger fear is not what AI might persuade me to do.

It is what someone with power, money, or simply a bad purpose may instruct AI to do.

There are at least two versions of that danger.

The first is familiar from science fiction. We create increasingly autonomous systems, give them objectives and access to the world, and eventually discover that human beings are no longer meaningfully in control.

That does not require a machine to become conscious, angry, ambitious, or evil. It does not have to hate us. It merely has to become sufficiently capable of pursuing an objective that our ability to supervise, understand, redirect, or stop it becomes inadequate.

But there is a second possibility that seems to me even more troubling.

What if the AI remains perfectly obedient?

Imagine that I own a candy company. I give an extraordinarily capable AI an assignment like this:

Maximize profit. Stay within the law, or at least avoid conduct likely to result in successful prosecution. Evaluate regulatory, civil, and reputational risks and operate as close to those boundaries as profitability justifies. Avoid scandal. Keep sensitive methods confidential. At the same time, maintain the appearance of a socially responsible company concerned about nutrition, children, workers, communities, and the environment.

Then I turn it loose.

I have not told the machine to poison anybody. I have not ordered it to commit a crime. I have simply omitted morality from the assignment.

What might a sufficiently capable machine discover?

Perhaps it finds formulations that encourage greater consumption. Perhaps it identifies the consumers most susceptible to particular forms of advertising. It may discover that an obscure disclosure satisfies the letter of a regulation while leaving most customers with a different impression. It might calculate whether reformulating a product is cheaper than lobbying to change the regulation governing it. It could determine precisely how much should be contributed to public-health programs to produce the greatest reputational benefit at the lowest cost.

None of this is new. Tobacco companies, food companies, financial institutions, technology companies, and other large organizations have spent generations learning how to reconcile profit, regulation, public relations, and varying degrees of conscience. We did not need artificial intelligence to invent greed, deception, regulatory gamesmanship, or moral flexibility.

Nor is every corporation corrupt. The point is that human organizations contain something that a machine optimized for efficiency might regard as a defect, friction.

Someone objects. Someone refuses. Lawyers become nervous. Executives disagree. Employees gossip. A whistleblower copies the documents. Somebody has a conscience. Somebody makes a mistake. Somebody goes home and tells a spouse what happened at work and hears, “You did what?”

From the perspective of pure optimization, these are inefficiencies.

From the perspective of civilization, some of them are safeguards.

Now imagine replacing portions of that organization with artificial agents capable of examining millions of transactions, customers, regulations, advertisements, prices, scientific papers, lawsuits, and political environments simultaneously. They never sleep. They do not become embarrassed. They do not need promotions. They do not worry about what their children will think of them. They can test thousands of strategies, retain what works, discard what does not, and continuously refine the successful ones.

The danger is no longer one unscrupulous executive or even a roomful of them.

It is the possibility of industrializing amoral optimization.

And still the machine has not rebelled.

It has obeyed perfectly.

That changes the question we should be asking about artificial intelligence.

“Will it obey us?” is not enough.

We must also ask, obey whom? Pursue what objective? Under what limitations? With what transparency? Who can stop it? And who remains responsible for what it does?

Which brings me back to the surgeon at the console.

We do not consider robotic surgery safe merely because the machine faithfully follows the surgeon’s movements. The surgeon is trained. The surgeon is licensed. The hospital and equipment operate within systems of standards and accountability. Failures can be investigated. Responsibility has not been transferred to the robot simply because the robot performs much of the physical work.

The machine is embedded in a human system of responsibility.

Artificial intelligence will require something analogous, although the problem is vastly larger. As these systems acquire greater capability and autonomy, the question will not simply be whether they are intelligent or whether they follow instructions. It will be whether human beings remain meaningfully responsible for the purposes to which that intelligence is directed, and whether we build systems that can hold the person at the console accountable.

And now we return to this essay.

Artificial intelligence wrote much of it. I want to say that plainly because concealing it would undermine the point I am trying to make.

The ideas did not begin with an instruction to “write me an essay about AI.” They developed in conversation. I supplied experiences, concerns, examples, objections, and the principle of being cooperative without being co-opted. A separate conversation about robotic surgery supplied the image of the surgeon at the console. I asked AI to research how that surgery works. AI recognized that the image belonged with the argument we had already been developing. We worked the two conversations together.

Then I told the machine what I wanted the essay to say and asked it to do a substantial part of the writing.

That is not a confession. It is the experiment.

I will edit what it produced. I will reject things. I will change sentences that sound too much like AI and sentences that simply do not sound like me. I may delete an elegant paragraph because it makes a point I do not believe. I may restore something the machine considers unnecessary because it matters to me. I decide what survives.

I am responsible for the finished essay.

Does that make me its author?

I think it does.

But it would be equally dishonest to pretend that AI was merely a fancy spellchecker. It contributed language, structure, research, connections, and ideas. This particular essay could not have been written in quite this way without it, any more than the surgeon could have performed precisely the same operation with unaided hands.

That is what makes the technology so exciting.

And that is what makes it frightening.

We are building instruments capable of magnifying human intellectual power, much as robotic surgery magnifies the precision and reach of a surgeon’s hands. Properly used, they may allow ordinary human beings to do extraordinary things.

But magnification is morally neutral.

It can magnify wisdom or foolishness, generosity or greed, careful judgment or an impoverished objective pursued with breathtaking efficiency.

So perhaps the question I want to carry into the age of artificial intelligence is simpler than all the technical ones.

Not, is the machine intelligent?

Not even, is the machine obedient?

Ask instead, who is at the console?

And then ask the harder question.

What has the person at the console told it to do?

Let the Machines Do the Work

At the North Oakland Senior Center the other day, a group of us fell into the very common conversation about AI.  A couple of us actually use it.  We like it; we're cautious but optimistic.  Most of the other Seniors were uneasy.  They worried about jobs disappearing, about machines replacing human contact, about a world where you press buttons and never reach a person, where machines make decisions affecting our lives.  

The anxiety wasn’t abstract. It was personal, children, grandchildren, the future. AI is not the first time we’ve had fears like these. When television arrived in our homes, when calculators appeared in classrooms, when computers entered offices, when cellphones became universal, people said the same thing.  This will ruin us. Children wouldn’t be able to think. We won’t be able to concentrate. People today will lose basic skills. Our minds will turn to mush or worse yet our children’s minds. 

AI is different in one important way. It isn’t replacing muscle or just making an existing task faster. It is beginning to do routine brain work; work that required clerks and procedural work by professionals: entering information, sorting it, finding it, organizing it, producing reports. It can write routine documents, compare information, and interpret it. The changes won’t stop at the bottom. They are already reaching into the cubicles of lawyers, accountants, the examining rooms of doctors, and the work of radiologists, any job that requires routine analysis. 

During my transition from banker to Park Ranger I got to work as a temp in Morrison and Foerster’s mail room.  On my mail rounds I passed cell like offices of associates who spent their days cutting and pasting boilerplate language for briefs and contracts. It reminded me of medieval copyists. Much of that work is already automated. AI will finish the job once it is made foolproof from fictional citations.We have been using technology for centuries to do more work with less labor. Computers eliminated filing cabinets and much of the clerical work force. Every time we make jobs more productive, we create the possibility of having more of what we need with less work.

The question isn't whether AI will eliminate certain kinds of work.  It will.  The question is what we do with our time and the increase in productivity after those jobs are gone.  When we buy a dishwasher do we complain about the time we no longer spend washing dishes.  No, we use that time to do things we enjoy; watch TV, read a book, do macrame or maybe more time at our job.  

For a number of decades now it doesn’t seem like we’ve been using that saved time to live better.  We’ve been using it to produce more, consume more, work more, and make more money for those of us with the new skills required. We were once told that technological progress would give us leisure. Instead, we found ways to make more products working harder and longer. 

AI today gives us an opportunity to make that choice again.  To revisit the question we haven’t really addressed adequately.  What to do with our new productivity?

Where machines can handle more of the basic work of feeding, clothing, housing and organizing schedules, increasing productivity by doing more work with less labor, we create more profit from the same amount of human effort. We can spend that increased wealth on doing other things. It is a remarkable opportunity. If it takes less actual work to provide the things we need, we have more time and resources for things that machines cannot give us: caring for elders, teaching children, restoring wetlands, maintaining public spaces and parks, supporting communities, cooking, making things with our hands, making art, writing poems, welcoming strangers, comforting someone who is frightened.

Some years ago I got to be an EMT in a State Park.  When there was a road accident, I showed up.  The first thing I did was tell the patient “You’re going to be alright,” and then I gave them what first aid I could and got them ready to go the hospital. Then we waited for the distant ambulance to arrive.  Sometimes I would hold their hand or put my hand on their shoulder. Today AI can assess the situation, recommend treatment, come up with a plan, and alert an ambulance.  But AI can’t put it’s hand on someone’s shoulder or hold someone’s hand. 

The promise of AI should be to free us from work that machines do better, not to eliminate humans but to redirect our efforts to things that only humans can do.

The movie industry is especially anxious about AI, perhaps because AI can already produce the same formulaic, explosion‑heavy content that studios churn out. But that only proves the point: AI threatens the mechanical parts of creativity, not the human ones. The great films, the ones with real imagination, aren’t formulaic, and they aren’t threatened. AI can clear away the tedium, but it won’t replace inspiration.

The gains from increased productivity aren’t shared.  If AI eliminates 10 jobs that used to be required, the savings seem to go to the few, the owners, instead of the people making the sacrifice.  AI can be used to make life better for everyone.

It’s a political decision. I come from a long line of wage laborers. I’m not opposed to private enterprise. Ownership and markets drive innovation and make economies more efficient. But productivity should strengthen the Commonwealth, not just the ownership class. If the gains from technology are concentrated at the top while everyone else is told to work harder to keep up, we have used innovation to make a few people extraordinarily wealthy rather than to improve the quality of life for all of us.

We can imagine something better.

The Commonwealth is more than the economy. It is the quality of the life we share. If technology gives us more productive capacity and more time, we can choose to live better rather than simply produce more.  

Yes, I’m idealistic. But if we don’t have ideals, what do we have?

AI is here and more is coming. We cannot stop it, and I don’t think we should want to. The question is who is in charge of what happens next?  Corporations? Investors?  Or the voters.

Productivity should serve life. Innovation should raise the floor for everyone. If machines can do more of the work required to sustain us, then the measure of progress should not be how much harder we work. It should be how much better we can make life for all of us. 

 

 

This essay was written with AI assistance. Editing is like sanding. I used an electric sander to get there and then finished it by hand.