There is No "We" in the AI Revolution

Forgetting what a human being is for while chasing the finish line

Pieter Bruegel the Elder – The Tower of Babel (1563)

It has been fascinating watching the rapid movement of the “AI Revolution” and then listening to the prophecies of where it all leads. The truth is, nobody really knows. In the short term, I think it’s undeniable that there will be significant challenges in the labor market. I am concerned about what it does to our willingness and ability to think and reason as a people. I am grateful for the many ways I have found to use AI in my own life and work, which have made me more efficient and productive. I am highly skeptical of the claims that it will make all of us rich. Like many of the technologies that came before it, AI is certainly a game-changer in many ways. But we are continually being told that we have to “keep up” and beat other nations in capabilities and advancement instead of slowing down and asking basic, important questions.

There are several prominent voices now commenting that AI engineers have long asked whether they could do things without asking the more important question of whether they should. It is a fair question, but the insinuation that it hasn’t been asked at all misses the point.

In 1960, Norbert Wiener argued that a machine which learns will pursue the goal we actually give it to pursue rather than the goal we meant, and then we can only notice the difference after the machine has acted on it. In a 1948 study, Wiener coined the term Cybernetics, or the scientific study of control and communication processes shared by living organisms and machines. His focus was systems that regulate themselves through feedback. On a small scale, if you have ever asked an AI agent to create an image you had in your mind and got something back completely different, you understand a basic principle of his argument. In other words, there’s no way to know what you’re going to get until after it is done and it depends entirely on whether or not you understand the input the machine wants in order to give you the outcome you want.

Joseph Weizenbaum created ELIZA at MIT in the mid-1960s, one of the earliest programs capable of holding a conversation with a person. It worked by picking up on keywords in what someone typed and turning them back into questions or comments like a therapist who just mirrors what the patient says. Weizenbaum’s intent was to demonstrate how shallow this kind of exchange really was. Instead, knowing full well it was only a program, his own secretary once asked him to leave the room so she could talk to ELIZA privately. Ten years later, he published Computer Power and Human Reason to argue that certain judgments should never be handed to a machine, no matter how convincingly the machine might seem able to make them.

Hans Jonas published The Imperative of Responsibility in 1979 with the simple claim that the ethics we inherited were built for a smaller world. In that older world, what people did mostly affected those nearby, and the results showed up within a lifetime, but modern technology has changed the scale. He pointed to things like nuclear power, industrial pollution, and genetic engineering. Jonas’ main point was that our old moral rules no longer fit this new reality, and that ethics needed to shift its priorities toward long-term responsibility.

C. S. Lewis published The Abolition of Man in 1943, before the first stored-program computer was ever booted up. He warned that if we treat all moral values as mere personal feelings rather than real truths, the people who control education and technology will eventually gain the power to reshape human nature itself. Once that happens, a small group of conditioners can decide what future generations should want and turn the rest of us into products of their design. The final result would be the end of anything recognizably human.

Aldous Huxley published Brave New World in 1932. In Huxley’s world, there is no need to ban books or burn them, because nobody reads them. The population is kept content through pleasure and a drug called soma that dissolves any discomfort. Huxley’s warning was that we would hand our freedom over to a tyrant in exchange for comfort.

In 1949, George Orwell published 1984 and imagined the reverse of Huxley. Here the control is total and unmistakable. Society is a panopticon under constant surveillance by a Party that manufactures its own truth, and a language called “Newspeak.” In time, the things that cannot be said will fall out of the thinking of the people. Orwell feared a world where information is withheld, and with that, a rewriting of history to serve those in power.

Ray Bradbury published Fahrenheit 451 in 1953, and though it is remembered as a story about burning books, Bradbury insisted the fire was never the real danger. In his telling, the people had stopped reading books long before the state began to burn them. They preferred the wall-sized screens with the sound piped constantly into their ears. Censorship was the final step.

In Amusing Ourselves to Death, Neil Postman argued in 1985 that our age was drifting toward Huxley’s world rather than Orwell’s. The greatest threat he saw was what we love. The instrument of ruin would arrive as a source of endless entertainment, and we would embrace the very things that eroded our capacity to reason. Later, in Technopoly, he argued one of the most important questions about any new technology is what it undoes.

So the “should we” question has been asked repeatedly for several decades. Weizenbaum is assigned in computer science ethics courses all the time. Jonas shaped German bioethics law. So what is really going on?

The question requires a telos

Telos is the inherent purpose, goal, or final end toward which a thing naturally aims or develops. Aristotle explained how you cannot judge whether a flute player is good without knowing what flute playing is for, and you cannot judge whether a man’s life is good without knowing what a man is for. Goodness is relative to function. If you ask the members of a corporate ethics board a telos question, you are likely going to get answers that are all measurable. They will talk about demographic bias, potential misuse, benchmark performance, privacy exposure, or downstream labor effects, but they will not talk about whether a human being is better off because of their company or products. That doesn’t really matter, and to answer it requires knowing what human beings are for. It is a question about human purpose and where meaning comes from, which doesn’t really matter in a technology company.

The “should we” question is easily reduced to nothing more than a risk assessment. If there is a risk, the focus shifts to mitigation, which is designed to modify a product rather than eliminate it. It may slow down production for a time so work can be done on new ways to deliver the product as safely as possible. There will be a cost/benefit analysis for investors, but asking a company to say “we will not do this” on principle because we really are concerned about the telos is like speaking Chinese to an Alaskan Salmon.

Many risk management offices and ethics boards are filled with well-meaning people who take harm seriously. But they also know saying “never” is not an option. They may want to build something a little different, or make a small change here or an adjustment there, but cutting it altogether because it is not ultimately good for humanity and the downstream effects could be disastrous won’t happen.

There is no “we”

Lewis said that what we call man’s power over nature turns out to be the power of some men over other men, with nature as their instrument, and that the group holding the power grows smaller with each generation of conditioners. That is the flaw hiding inside the question “should we build this?” It assumes a “we” that can weigh the matter and say no. But if Lewis is right, there is no such “we.” Do you really want our future decided entirely by Sam Altman, Dario Amodei, and Elon Musk? Do we trust them enough to give them that kind of power?

The same logic scales from the firm to the nation. We are told the real question is who will be the first to build? If we hesitate, China will not, and so restraint on our part is surrender. It’s an argument from fear. Fear of losing the race, so we need to keep running. We will deal with any consequences later.

No doubt, the competition is real. No doubt, everyone is rightly concerned about who holds the power. But the argument takes the one question worth asking, whether we should build this at all, and replaces it with whether they or we should be the ones to build it. In other words, the second question answers the first, and once we frame it that way, “should we?” no longer has a place to stand, because every objection can be met by pointing across the ocean. The acceleration becomes justification, so how can we ever make it stop? It lets a company present its commercial ambition as patriotic duty. Shouldn’t we all be scared?

And suppose “we” win the race. What have we actually won? Keeping up says nothing about the kind of people we become while we do it. Are we desperate for more weapons of war? Do we need more sophisticated surveillance? Are we meeting actual human needs, or are we handing the tools of the West’s downfall to firms with different priorities than civil liberties and individual well-being? When our focus is on winning the race, we get to a finish line we never agreed to or asked for.

What good are the universities?

Very little research is required to find that the study of humanities in most universities has been gutted. Sixty years of professional humanities scholarship went into establishing that there is no fixed human nature to appeal to. Out with Plato, Aristophanes, Dante, Milton, Shakespeare, and Tolstoy. And then, out with humanities altogether. When West Virginia University eliminated 28 academic programs and 143 faculty positions in September 2023, including every foreign language major on the Morgantown campus, the cuts were shaped by a state funding formula keyed to workforce development. The programs closed, and no measurable capacity was lost. When the institutions serve their students Flaming Hot Cheetos, don’t expect them, in turn, to deliver Michelin-starred Curry. It is easy to blame the robotics lab and let the English and philosophy departments play victim. But what are the remaining humanities departments even doing? There is plenty to choose from. One PhD dissertation from Cambridge University is entitled: “Exploring gendered embodiment in transgender and gender-diverse adolescents at the intersection of exercise and eating behaviours, social inclusion, and mental health: Towards a desire-centred approach in paediatric gender-affirming healthcare and research.” Delicious Flaming Hot Cheetos for breakfast, lunch, and dinner. No thank you.

We can ask “should we?” all day, but who gets to answer the question? I hope it’s not the illustrious scholars at Cambridge or any of the other “most prestigious” universities from around the world.

Cain’s city and Bezalel’s workshop

In Genesis 4, Cain builds a city and names it for his son (4:17). His descendants invent animal husbandry, the lyre and pipe, and instruments of bronze and iron (4:20-22). Four verses later, Lamech tells his two wives that he has killed a man for wounding him, and that if Cain’s avenger came sevenfold his own comes seventy-sevenfold (4:23-24). In the biblical text, there is no interval between a new capability and its use against a neighbor.

Exodus 31 reveals the other side. The Lord tells Moses that he has called Bezalel by name and filled him with the Spirit of God, “with ability and intelligence, with knowledge and all craftsmanship” (31:3). His skills are a gift. The object is specified by someone other than the craftsman, item by item, down to the dimensions of the Ark of the Covenant and the composition of the anointing oil. Bezalel is a superb workman with a specific task and a noble end. A desirable end must be given from outside the work. Engineers inside the project can’t reliably deliver the end. An ethics board within a firm won’t act with pure motives, despite good intentions. What remains are institutions that operate without ever actually wondering what man is for. There may be a momentary pause in thinking, but there’s a race to be “won” and a product to deliver, so the real moral questions are quickly suppressed.

What man is for

I don’t have five points that would slow the race down, and I am suspicious of anyone who says they do. Cain built his city, and his descendants turned their new tools against a neighbor. Nothing in the centuries since suggests we have even considered leaving a larger interval between capabilities and their large-scale use. To say that a technology might not be good for a human being, you have to believe there is such a thing as a human being, and that a human being is for something. Bezalel’s skill was a gift, and his work was good because God handed the end to him, and only God has the standing to determine the desired end.

So the task before ordinary people is to remain the kind of people who still know what a human being is for, when nearly every institution around us has decided to ignore it. We still have real agency. We decide how we use the tools and what we refuse to hand over to them. We keep thinking, while we are being trained not to, refuse the trade of freedom for comfort that Huxley predicted, and hold the conviction that we are not raw material for someone else’s design. The ability to think slowly and carefully is not a technical skill. They are principles and habits that can be worn down, but we can work every day to hold onto them, we don’t have to give up on them. Nobody knows where the AI revolution leads, but we don’t have to keep up in the race to know what we are for. The old questions are still the ones worth asking.