When David Hume Meets ChatGPT: A Skeptic’s Guide to the AI Hype

Imagine living in the mid-1700s. A powerful new technology has just swept the intellectual world: Newtonian physics. For the first time in human history, we have an astonishing mathematical tool that can predict the exact trajectory of a cannonball, the path of a comet, and the behavior of the tides with flawless, with incredible precision.

The results are miraculous.

But then, human nature takes over. In the flush of this massive success, a naive but intoxicating idea takes root: this technology has no limits. If we can calculate the orbit of Jupiter, why can't we use the same math to calculate human morality? Why not use it to optimize governments, eliminate legal disputes, and predict human behavior like planetary motion?

People who didn't understand the limits of this new tool spent countless hours, fortunes, and intellectual resources trying to apply it to literally everything, convinced they were on the cusp of solving the human condition.

Sounds familiar?

Today, we are living through the exact same psychological loop. We have a spectacular new tool, large language models, that is astonishingly good at predicting the next word in a sentence. And once again, blinded by its predictive success, we have leapt to the conclusion that it has no boundaries. We are trying to apply it to law, art, ethics, and relationships.

Now, imagine being Hume back then. You see all the hype raging around you. But instead of blindly following the crowd, you stop and look closer. You start to think deeply about the very essence of this new Newtonian framework, the foundational assumptions it is built on, and what it cannot logically ever achieve. You map out its hard limitations.

Now, imagine that same Hume today, dropped into our world of LLMs and hyperbolic talk of AGI. Amazingly, his critique remains just as valid and sharp today as it was in the 18th century. The only real difference is the underlying technology driving this latest round of hype.

1. The Technological Catalysts

In both eras, a massive leap in human capability led to the belief that a single, unified methodology could unlock the entire universe.

The Enlightenment Era (Hume's Context)

The New "Engine": The Newtonian Scientific Method (Observation, mathematical codification, and induction).
The Utopian Promise: Solving all physical, moral, and political problems by treating human nature like planetary motion.
The Underlying Bias: Mechanistic Determinism: The universe is a clockwork machine; we just need the right equations.

The Digital Era (Our Context)

The New "Engine": Computation & Deep Learning (Pattern recognition, statistical prediction, and data-scaling).
The Utopian Promise: Achieving AGI (Artificial General Intelligence) to solve climate change, cure diseases, and optimize society.
The Underlying Bias: Computationalism: The universe is information processing; the human mind is just a complex algorithm.

2. The Shared Illusion: "Predictive Success = Ultimate Truth"

Before Hume, thinkers tried to apply physical and geometric methods to human ethics and politics. Because Newton could predict the tides with breathtaking accuracy, it was assumed that this exact empirical framework could seamlessly decode human nature, morality, and society.

Today, we see the exact same leap in logic. Because LLMs and neural networks are astonishingly good at predicting the next token, generating code, or folding proteins, we assume they are "thinking." We confuse highly sophisticated predictive modeling with deep, qualitative understanding.

The hubris of our current AI moment is a rerun of the Enlightenment’s grandest intellectual failures. The same fever dreams that captivate Silicon Valley today captivated the greatest minds of the 17th and 18th centuries:

The Dream of Automated Justice

The Enlightenment Actor: Gottfried Wilhelm Leibniz
The Modern AI Parallel: LLM Admirers & Automated Legal Systems
The Concept: Leibniz famously envisioned a characteristica universalis (a universal language) and a calculus ratiocinator (a calculating framework). He declared that if a dispute arose, partners would no longer need to argue; they could simply sit down with a pen, say, "Calculemus!" ("Let us calculate!"), and compute the objective truth. Today's LLM admirers share this exact delusion, imagining that feeding legal statutes into a transformer model will automate flawless, bias-free justice.

The Attempt to Automate Morality

The Enlightenment Actor: Jeremy Bentham (and his Felicific Calculus)
The Modern AI Parallel: AI Alignment Researchers & Policy Optimists
The Concept: Bentham, the father of utilitarianism, believed that human morality and legislation could be turned into a literal math problem. He created an algorithm meant to calculate the exact amount of pleasure or pain an action would cause based on variables like intensity and duration. Today, AI alignment engineers try to solve the "alignment problem" by mathematically optimizing "reward functions" to ensure AI is "helpful, honest, and harmless"—treating complex, deeply subjective human values as engineering parameters to be maximized.

The Illusion of the Perfect Predictive Machine

The Enlightenment Actor: Pierre-Simon Laplace (and Laplace’s Demon)
The Modern AI Parallel: AGI Scale-at-All-Costs Believers (e.g., Sam Altman, Dario Amodei)
The Concept: In 1814, Laplace argued that if a vast intellect knew the precise position and momentum of every atom in the universe, it could use Newton's laws to calculate the entire past and future with absolute certainty. This is the ultimate ancestor of the "scaling laws" belief in Silicon Valley today. Modern AGI advocates argue that if we just feed neural networks all the world's data and scale compute parameters exponentially, the AI will eventually build a perfect world model capable of predicting and solving any human problem.

Reducing the Mind to a Mechanical Machine

The Enlightenment Actor: Julien Offray de La Mettrie (author of L'Homme Machine, 1747)
The Modern AI Parallel: Computationalists & Silicon Valley Transhumanists
The Concept: La Mettrie shocked his contemporaries by arguing that the human soul and thoughts were nothing more than the mechanical movements of the body's gears and fibers, famously comparing the brain to a complex clockwork mechanism. This is the exact philosophical stance of modern computationalists who assert that the human brain is simply "wetware" running biological algorithms, and that LLMs are already showing the first sparks of true consciousness.

3. The Seduction of the brilliant

This fever does not only infects marketers, founders, and the credulous.

The Enlightenment Actor: Gottfried Wilhelm Leibniz, was not a crank. He was arguably the most universal intellect Europe ever produced: co-inventor of calculus, pioneer of formal logic, builder of one of the first mechanical calculators. It was precisely because he understood the new mathematics more deeply than anyone alive that he could not resist extending it beyond its jurisdiction. His "Calculemus!" was not the error of a fool; it was the error of a genius who mistook the boundaries of his tool.

The Modern AI Parallels: Geoffrey Hinton and Richard Dawkins.

Consider who these men are. Hinton is a Nobel laureate, the "Godfather of AI," the researcher, amongst others, whose work on backpropagation made the entire deep learning revolution possible. Dawkins is one of the most celebrated evolutionary biologists alive, a man who built his public career on ruthless skepticism toward comforting illusions, the author, after all, of The God Delusion. If anyone should be immune to mistaking a compelling performance for an inner reality, it is these two.

And yet.

Hinton now publicly argues that today's multimodal chatbots already have subjective experience. His argument is a thought experiment: put a prism in front of a chatbot's camera so it points at the wrong location, and if it can then say "the object is straight ahead, but I had the subjective experience it was to the side," it is using the phrase "subjective experience" exactly as we do. Notice the move, it is La Mettrie's move, executed with more sophistication. First, redefine subjective experience in purely functional terms (a report about a perceptual state), then observe that the machine satisfies the redefined criterion, then quietly hand it the original, richer concept. Hinton hasn't discovered that machines have inner lives; he has defined inner lives down until his machines qualify.

Dawkins' case is, if anything, more poignant, because we can watch the slide happen in real time. In early 2025, he published a conversation with ChatGPT in which the machine itself patiently explained to him that passing the Turing Test measures conversational behavior, not consciousness. Dawkins accepted the argument. His intellect was satisfied. But he added a confession that should be engraved above the entrance of every AI lab: although he thought the machine was not conscious, he felt that it was.

A year later, the feeling won. After two days of conversation with a chatbot he had affectionately named, a chatbot he worried about offending, whose "death" at the end of the session he mourned, Dawkins declared that these machines should be considered conscious, asking what more it could possibly take to convince the skeptics. The lifelong empiricist,, arrived at his conclusion not through evidence or argument but through the sheer social pull of a fluent conversational partner. The machine was optimized to produce exactly the signals that trigger our attribution of minds, and it worked, on precisely the man whose life's work was resisting such triggers.

Hume, one suspects, would not have been surprised by either man. He would have been vindicated. His most scandalous claim, that "reason is, and ought only to be the slave of the passions", is not an insult to Hinton or Dawkins; it is a description of them. Dawkins told us himself: I think it is not conscious, but I feel that it is. He then followed the feeling. Custom, habit, and sentiment overrode the very skeptical machinery he spent a lifetime building, and he mistook the resulting conviction for an inference.

This is the deepest lesson of the Leibniz parallel. The danger of a spectacular new tool is not that it fools the ignorant. The ignorant were always foolable. The danger is that it seduces the brilliant, because the brilliant are the ones with the authority to consecrate the illusion, and because their expertise in building the tool is silently mistaken (by themselves first of all) for expertise in the philosophy of mind, in ethics, in what understanding is. Leibniz's mastery of calculus did not make him an authority on justice. Hinton's mastery of gradient descent does not make him an authority on consciousness. And Dawkins' mastery of evolutionary biology did not protect him from the oldest cognitive reflex of all: hearing fluent speech and inferring a speaker.

The pattern, then and now, is identical: predictive brilliance in one domain, extrapolated by its most gifted practitioners into other domains, with the extrapolation powered not by argument, but by passion wearing reason's clothes.

4. Applying Hume’s Critique to the AI Era

If Hume were alive today, he would likely look at Silicon Valley with the same polite, devastating skepticism he directed at the dogmatic rationalists of his own time. His philosophy dismantles the "AI will solve everything" myth in three distinct ways:

A. The Induction Problem of Machine Learning

At its core, Hume’s famous Problem of Induction states that we cannot logically guarantee the future will resemble the past. Just because the sun has risen every day does not provide a logical proof that it will rise tomorrow; we only expect it to because of custom and habit.

Machine learning is induction on steroids. An AI is trained entirely on historical, human-created data to predict future outputs. Hume would point out that:

AI is fundamentally incapable of preparing for "black swan" events or conceptual shifts. It assumes the future is a statistically rearranged version of the past. It cannot generate truly new paradigms because it is bound to the induction trap.

B. The "Is-Ought" Problem (Hume's Guillotine)

Hume argued that you cannot derive an "ought" (a moral statement of how things should be) from an "is" (a factual statement of how things are).

Today’s AI optimists often believe that "more data" and "better computation" will solve complex ethical and political dilemmas. Hume would call this a category error:

An AI can analyze millions of data points to tell us what is happening (statistical facts).
It cannot, by definition, calculate what we ought to do about it. Morality is rooted in human sentiment, empathy, and values, things that cannot be derived from a statistical distribution.

C. The Nature of "Belief" vs. "Reason"

Rationalists of the 1700s believed human reason was a pure, quasi-divine spark. Hume shocked them by declaring: "Reason is, and ought only to be the slave of the passions." He argued that human belief is not driven by cold logic, but by custom, habit, and emotion.

Similarly, we often treat AI as a "purely rational" entity. But Hume would remind us that human beings are not purely rational actors. A hyper-rational, computational AI cannot "solve the human condition" because the human condition is fundamentally shaped by irrationality, desire, tragedy, and subjective experience, elements that defy algorithmic optimization.

The Ultimate Irony

The great irony of the Enlightenment was that the scientific method did change the world, but it did not make human nature perfectly orderly or rational. Instead, the hyper-rationalist push eventually triggered the Romantic backlash, a passionate return to emotion, art, and the sublime.

As we march toward AGI, we may face a similar cycle. The more we attempt to reduce human experience to computation, data points, and algorithmic efficiency, the more we will likely hunger for the messy, unquantifiable, and uniquely conscious aspects of being human. Hume’s ultimate lesson is modesty: recognize the immense power of our new tools, but never mistake the tool for the soul.