Le Bon's Crowd and AI Products
Thinking About Human Connection Rather Than Technical Solutionism
On the recommendation of a friend, I recently read The Crowd: A Study of the Popular Mind (French original Psychologie des Foules), by French polymath Gustave Le Bon, published in 1895. It’s a remarkable book, filled with crucial lessons about how we should think about the psychology of the masses.1
I’m in the business of building AI-related products for the masses—both LearnerShape PlaylistBuilder (a simple and effective YouTube curation application) and the Safe and Responsible AI Information Hub (Saihub.info)—and there are clear lessons from The Crowd for my work, and that of others building and delivering AI products.
What Le Bon Says About Crowds
There is too much in The Crowd to consider in a blog (I encourage you to read the book yourself—it’s only 100 pages), but two observations seem necessary.
First, Le Bon is specific about what he means by “crowd”.
I was initially perplexed by Le Bon’s reference in his preface to “the extreme mental inferiority of crowds”, particularly because I felt that I had learned the opposite from James Surowiecki in The Wisdom of Crowds. The reason for this apparent contraction is that Le Bon and Surowiecki mean very different things by “crowd”.
For Le Bon, a crowd is a gathering of individuals (not necessarily in the same location) who form a “collective mind”:
“Under certain given circumstances, and only under those circumstances, an agglomeration of men presents new characteristics very different from those of the individuals composing it. The sentiments and ideas of all the persons in the gathering take one and the same direction, and their conscious personality vanishes. A collective mind is formed, doubtless transitory, but presenting very clearly defined characteristics.”
Surowiecki means something very different by “crowd”: “the conditions that are necessary for the crowd to be wise [are] diversity, independence, and a particular kind of decentralization”. That is, wise crowds are groups of independent individuals from whom information is gathered to support decisions (or for other reasons). Surowiecki’s work is important and very useful, but it addresses an entirely different phenomenon from Le Bon, and it’s Le Bon’s ideas that are the subject of this blog.
Second, the central thrust of The Crowd is that a crowd acts impulsively, according to collective beliefs and illusions, which have little necessary connection with objective reality, general morality, or the usual beliefs and personalities of the individuals forming the crowd.
[A]mong the special characteristics of crowds there are several—such as impulsiveness, irritability, incapacity to reason, the absence of judgment and of the critical spirit, the exaggeration of the sentiments, and others besides … .
…
A crowd is at the mercy of all external exciting causes, and reflects their incessant variations. It is the slave of the impulses which it receives. The isolated individual may be submitted to the same exciting causes as the man in a crowd, but as his brain shows him the inadvisability of yielding to them, he refrains from yielding.
Le Bon’s finds the influence of such irrational crowds to be pervasive in society, including not just mobs and popular demonstrations, but also more durable social groups, political groups, members of legislatures, juries and various others.
This narrative of a dominant role for emotionally-driven crowds has lacked currency in recent decades in our technocratic society. The historical books of political philosophy that are in the ascendant in recent decades tend to be ones focused on “strategy”, such as The Art of War by Sunzi or The Prince by Machiavelli. This not surprising: scientists, economists and consultants like to believe that strategy works—and sometimes it does. But Le Bon’s recognition that often-irrational emotion is hugely important in determining human behavior deserves greater attention.
Making a Human Connection with AI Products
So … you may be thinking … what does this have to do with AI products?
Rather a lot, in fact.
Essentially, if one believes that Le Bon is correct, then the prevailing AI solutionism advanced by Silicon Valley tech bros, and their high priests like Sam Altman and Marc Andreesen, is missing a few tricks. A dramatic statement of this worldview is in Andreesen’s Techno-Optimist Manifesto:
Combine technology and markets and you get what Nick Land has termed the techno-capital machine, the engine of perpetual material creation, growth, and abundance.
We believe the techno-capital machine of markets and innovation never ends, but instead spirals continuously upward.
I am a believer in technology—it’s what I do for a living—but it’s trivial to observe that the idea that the solutions to all or most problems are driven by technology and markets omits the human dimension. And if you believe what Le Bon has to say, this is a critical omission.
That is, where the behavior of human crowds is not rational (and that’s in virtually all spheres of life according to Le Bon), solutions may not be all of what we need. Even if we manage to convince crowds for the time being that the miracle of artificial intelligence will make their lives better, they may (like many crowds) change direction and opinions rapidly if and when they discover that their humanity has been co-opted by technology and markets.
I see two major lessons from Le Bon for marketing of AI products.
Don’t assume that people are looking for a solution.—The central lesson of The Crowd is that people are often not rational. Usually, they look for validation of emotional needs and illusions.
Think about the human dimension. Closely related to point 1, it is crucial to make an emotional connection with the humans who are users of or affected by an AI product.
If we don’t heed these lessons for AI products, we may be heading for social disruption, revolution, or worse. While it is easy for Silicon Valley to make the case that AI products will improve corporate profits through efficiency (and indeed the efficiencies of AI products hold great economic promise), the idea that we will simply put humans out of work, and support them with universal basic income and/or similar, is a road to perdition.
Recall the much-cited Luddite movement, involving destruction of automated looms by workers in English textile mills of the early 19th century. This is often given as an example of resistance to technology, but that misses the point. More complete descriptions of the movement recognize that the core problem was technological innovation that benefitted capitalists to the detriment of workers. People want the dignity of good work.
My ideas are above not revolutionary. Indeed, they are highly aligned to a basic tenet of start-up innovation: that it is crucial to address real needs. The broader point that I draw from Le Bon is that those needs must include human needs, or the crowd will bite back.
These ideas are also very contemporary, even though Le Bon wrote 130 years ago. Witness the growth in the last decade of popular movements around the world, especially on the right but also on the left. These are driven by the same phenomena that Le Bon identifies in The Crowd.
We in the AI industry, in the world’s technological vanguard, should pay attention.
There is much in The Crowd that is objectionable from a contemporary viewpoint, in particular its racism and sexism, but that does not significantly detract from the overall message. Readers should not assume that I endorse the book in every detail.