Companies Like AI Because It Allows Them to Work The Way They're Actually Designed To Work
Strong Opinions, Weekly Held
If your inbox and LinkedIn feed are at all like mine, they’re overcrowded with AI updates, predictions, and think-pieces. So I feel a bit squeamish writing this post.
But I have thoughts! Plus, I got some interesting reactions to my hypotheses that AI adoption will lag behind the technology and that organizations wanting to go all-in on AI might be better off firing everyone, so maybe my pontificating is novel enough to be worthwhile?
For those still intrigued…
I’ve been ruminating on why the rush toward AI has felt so different than other technology transformations of the past several decades. Yes, I do think the technology itself may offer a leap forward that’s not just evolutionary but genuinely revolutionary. That alone may account for the AI salivation. Fair enough.
But I have a different hypothesis that I want to begin exploring here. A darker, sociotechnical explanation that goes something like this: AI allows organizations to work as their actual design intended. Design which, for the better part of the last century, has been stretched to accommodate the messiness that is human labor.
Join me – 7pm at Brooklyn Brewery - tickets available here.
I’ve been thinking a lot about how organizations today are still built and run in ways remarkably similar to early, post–Industrial Revolution organizations. Think of the black-and-white photographs of hierarchical, chain-of-command, bureaucratic, assembly-line organizations. This underlying infrastructure continues to prevail despite the fact that most of today’s organizations have very little in common with the factories of the late 19th and early 20th centuries. A significant portion of modern organizations make products and services that didn’t exist a century ago. Heck, even the majority of jobs did not exist just 80 years ago, let alone further back.
Those old organizations were designed to use people like machines. Think: Frederick Taylor’s Scientific Management. Most of today’s organizations are still architected on those design principles, but find themselves dealing with two challenges: First, despite their structural DNA, many organizations don’t produce the assembly line widgets of those old companies. And second, partially as a result, modern organizations find themselves at odds with their human employees who don’t want to be treated like machine parts.
1910 photograph of the Samsun Tobacco Factory (Colorized)
On the other hand, today’s generative AI is fully compliant and doesn’t talk back. It conforms to the desired state of most companies: to run all inputs (including labor) as mechanized cogs in a machine. Thus, the promise of AI is that it offers capital the best of both worlds: machine-like compliance and output with humanesque creativity. If done right, AI should surpass the creativity, innovation, and critical thinking of the best humans without challenging power and authority. Frederick Taylor’s dream.
For most organizations, human labor is an inefficient, necessary evil. That’s why organizations have always replaced human labor with machines whenever possible. Even for more labor-friendly organizations, eventually the competitive marketplace comes calling.
(Side note: I don’t think that’s true for all organizations nor do I think it has to be. I definitely don’t think it ought to be. But as a seasoned organizational psychologist who’s observed hundreds of organizations, my strictly observational contention is that most actually function based on that premise despite what they otherwise espouse. And often, I don’t even think it’s intentional.)
Now, for the first time, the white-collar world of knowledge work has to contend with a technology that might give capital that magical combination of human performance and robotic compliance it’s been yearning for over the past century.
So a key question facing the future of organizations and AI is this: Is the messiness of creativity, innovation, etc. that’s necessary for a post-industrial highly specialized economy something that by its very nature, needs to be produced by humans? Are today’s products and services inextricably linked to the capabilities of a human mind? Or can AI replicate or mimic it well enough to obviate human labor?
I wish I knew. Maybe ask ChatGPT?
That’s it for this edition - please reach out if I can be at all helpful.
Be compassionate and deliberate.
I will let you know what Claude tells me :-)