View this post on the web at [link removed]

What does artificial intelligence have to do to be really, transformatively productive?
A lot of what it’s doing right now, or at least a lot of what the public sees, is not going to change the world. Writing fake articles and substituting for stock photos (which people used to use in their blog posts instead of AI-generated illustrations) is replacing work that is already relatively low-paid and not, alas, central to the economy.
More to the point, AI so far has been reducing rather than increasing the value of the kind of work we think of as requiring human creativity. The comment making the rounds on the internet, in various [ [link removed] ] forms [ [link removed] ], is that AI should be doing tedious tasks for creative people, but instead it’s doing creative tasks for tedious people.
With that in mind, perhaps the really transformative use of AI, the one that would free up large amounts of creative effort, is one that is largely being overlooked. What we need is the old dream of the household robot servant—an AI Jeeves who will fold your laundry and make sure coffee and breakfast are waiting for you in the morning.
Every Mother Is a Working Mother
One of the biases of Silicon Valley and the technology media is that they tend to focus on technology that involves the manipulation of information, which is their comfort zone, rather than the manipulation of material objects in the real world. They focus on bits instead of atoms [ [link removed] ]. John Deere, for example, has developed a self-driving tractor [ [link removed] ], but you don’t hear much about it, because the tech world isn’t paying attention to something so old-fashioned and literally dirt-covered as agriculture. They aren’t very focused on the household, either—but perhaps they should be.
We’ve already had a revolution in labor-saving household devices. According to a recent study [ [link removed] ] by the National Bureau of Economic Research, “Back in 1900, without household appliances, the average US household spent 58 hours per week on meal preparation, laundry, and cleaning.” As they used to say, every mother is a working mother. Despite the rosy conservative view of the traditional family, you can see why most women didn’t work outside the home. It wasn’t a luxury made possible by their husbands’ well-paying factory jobs. (In fact, most families enjoy a substantially higher standard of living [ [link removed] ] today.) Rather, it was a necessity due to the sheer volume of menial labor required inside the home.
The NBER report gives a description of just one of a housewife’s chores.
In 1900, around 98% of households relied on a 12-cent scrub board for laundering clothes. The process involved transporting water to the stove where it was heated using wood or coal. Clothes were then cleaned using either a washboard or a mechanical washing machine. Subsequent rinsing was necessary. To remove excess water, clothes were either wrung out manually or through a mechanical wringer. Then, they were hung on a clothesline for drying. The final step involved the laborious task of ironing, utilizing weighty flatirons that required constant heating on the stove.
When this description uses the term “mechanical,” bear in mind that no electric motors were involved. The machines were cranked by hand—almost exclusively a woman’s hand.
Then in the 20th century, a wave of innovation drastically reduced this workload: electric ovens, dishwashers, washing machines, vacuum cleaners and so on. Also, more subtly, an army of “home economists” performed time and motion studies and made calculations to determine the optimal height for sinks and the most efficient designs for modern kitchens. The results, again from the NBER study: “The time spent on household chores has declined sharply over the 20th century, reaching just 18 hours per week by 1975.”
By the mid-20th century, the reduction of unpaid labor in the home allowed women to migrate into the paid workforce. Women had been doing plenty of work all along, but now it was counted in the official economic figures because it was happening outside the home. The increased economic productivity is in large part a measure of how much labor was not being counted before and is now liberated for other uses.
And yet, “This process ... stabilized after the 1990s.” By the end of the 20th century, the revolution of household appliances reached a plateau. It has achieved all it can achieve. What we need is the next step.
An Army of Robot Servants
Combine this observation about women’s work with its effect on “men’s” work, which is that they tend to even each other out. Derek Thompson sums it up [ [link removed] ] in The Atlantic:
In the 1880s, when men worked long days and women were mostly cut off from the workforce, the typical American married couple averaged just over 68 hours of weekly paid labor. In 1965, as men’s workdays contracted and women poured into the workforce, the typical American married couple averaged 67 hours of weekly paid labor—just one hour less. In the early 2000s, the typical American married couple averaged, you guessed it, almost exactly 67 hours of weekly paid labor. In 2020? Still 67 hours.
The constancy of this figure implies that there is a certain natural limit to the amount of labor available within a family, and when one half of a couple works more outside the home, the other will work less to compensate. But it also implies that the labor required for care of the home also remains a constant. What if that could be further reduced?
It can’t be done with mere mechanical devices, because we already have those. Household machines have taken away or simplified much of the brute physical work. What remains is tending to the machines: gathering the dishes, sorting the clothes, folding them when they’re dry, preparing the ingredients to be cooked and so on. It’s not a huge amount of labor, but it is a significant expenditure of valuable time, and it can’t be automated mechanically. It requires intelligence and judgment—or something like human intelligence.
The idea of an artificially intelligent robot butler—a Jeeves to our Wooster [ [link removed] ]—is still a long way off. What we have now are the first incremental bits, like a robot vacuum that will help keep your floors clean but is just learning how not to smear around dog vomit [ [link removed] ]. There are some early attempts to devise a robot that can fold laundry [ [link removed] ], but it can’t yet match up your socks.
If having an AI-driven robotic servant is like having a butler, a self-driving car is like having a chauffeur, something that would reclaim many hours a week for most of us. Yet the self-driving car, which was definitely supposed to be here by 2018, has taken a lot longer than expected. Some are already on the road, but they only really work in two applications: robotaxis operating at low speeds on the well-mapped streets of a few city centers, and John Deere’s autonomous tractor, driving itself around in open fields where there’s not that much to crash into. Yet this is actually a less complex task than, say, cooking a meal.
It is a well-known rule of AI that tasks that seem incredibly difficult for humans, such as beating a chess grandmaster, are much easier for computers than tasks that are easy for humans, such as putting a kettle on to boil and making a cup of tea.
Interestingly, a group of executives from Cruise, one of the robotaxi companies, has just raised money for a new startup [ [link removed] ] devoted to “household robots.” It is unlikely we will get the robot butler of science fiction, a humanoid automaton that does all our tedious domestic tasks itself. But that’s as unimaginative as expecting our robot butler to operate an old-fashioned hand-crank clothes wringer. We tend to make the mistake of assuming machines will do things the way we do them, rather than the way that is easiest for a machine to do them. It is more likely we will gradually get a variety of appliances or specialized robots that perform specific tasks. One to gather, sort, clean, fold and return your laundry; another to gather and clean dishes and put them back on the shelves; another to take ingredients from the fridge and make breakfast; another to mow the lawn and weed the flower beds. And so on.
The Agatha Christie Conundrum
Like I said, this technology is a long way off—perhaps as far away as our contemporary household appliances were in 1900. The point in thinking about this is to project the value that would flow from having AI enable some kind of robots to perform these routine tasks.
It is well known, for example, that Western Europeans tend to spend fewer hours in the workplace than Americans. This is usually taken as evidence that Europeans have better priorities, valuing leisure time over work. But a recent study [ [link removed] ] by the Dallas Fed found that most of that extra time is not leisure. It’s spent in “home production,” i.e., housework. This, in turn, is because Europeans pay much higher taxes, so they don’t have money left over for buying time-saving appliances or acquiring household goods in the marketplace. They end up having to do things themselves at home. How could we escape this fate?
How did people do it before? If it used to take 58 hours a week of housework to support a family, how did anyone get anything done? Well, it was ordinary for a middle-class family to hire human servants to do household work. In her autobiography, the mystery writer Agatha Christie recalled [ [link removed] ] that in 1919, she and her husband, who were living on a salary the equivalent of about $50,000 a year today, thought it would be a foolish extravagance to buy a car. But for surprisingly little money, they had two full-time servants, who were “considered essentials of life in those days.” The servants were doing the 58 hours of work so Christie could have time to write her novels.
Since then, cars have gotten a lot cheaper, and for precisely that reason, human labor has gotten a lot more expensive. In a middle-class nation that is quickly becoming an upper-middle-class nation [ [link removed] ], we are all so expensive that we cannot afford one another.
(I suppose we could import an army of poor immigrants who would live in relative squalor while doing our household tasks for us. You could argue that it is only because we are so much richer than the rest of the world that we have the luxury of being shocked by this. Nevertheless, I don’t think there’s going to be much appetite for this solution.)
Suppose that eventually we are able to invert Agatha Christie’s economic calculation, and finding people too expensive to hire, we buy machines instead—but the machines serve the same function as household servants. What would that do for us?
The ‘Downton Abbey’ Effect
It’s possible that an army of robot servants would save us that final 18 hours a week of housework, which we could then free up for more paid work outside the home. It would certainly have the effect of helping to equalize working hours for women, who still tend to do the majority of household tasks [ [link removed] ].
Yet the data indicate that Americans already work a lot. And the idea that if we had less housework to do, we could put it all into “the grind” reflects that same bias in which anything that isn’t the traditional Silicon Valley economy—a bunch of young people pulling all-nighters on a startup—isn’t considered real economic activity.
The value of a robotic leap forward isn’t necessarily the other work it would free us up to do, but rather the value of the household work itself, all done automatically, without much thought or effort on our part—which implies a transformation in our quality of life.
In short, we could all live like aristocrats.
It’s important to remember that when we get a transformational new technology, we never use it simply to do exactly what we did before. When we got automobiles, we did not use them to travel the same distances we used to cover by horse or on foot. On average, people travel many more miles in a year than we used to, because we can. Think how we could live if we weren’t just doing the same amount of housework as before, only with less effort. Think of the extra housework we would demand if we didn’t have to do it ourselves.
Here’s just one small example. One of the most notable differences between Agatha Christie’s early 20th-century heyday and our own is the great casualization of our personal wardrobe—the baggy rumple of our jeans-and-t-shirts modern style, a world where Mark Zuckerberg can wear a hoodie [ [link removed] ] to a meeting with Wall Street investors and get away with it. There are many factors behind this, but surely one of them is economic. A wardrobe of smartly pressed suits takes time, and our time has become more valuable. Most of us no longer want to put in the long hours at the ironing board—but what if we all had an AI Jeeves to do it for us?
In effect, the rise of AI servants would mean that in theory, we could all live like characters in “Downton Abbey,” if we wanted to. It’s the kind of lifestyle where a servant helps you don an evening dress before a gong rings for dinner at eight, which is served to you at an elaborate place setting with cloth napkins. And why not, if the robots are doing all the work?
Of course, you don’t have to live like this, and many of us would continue to perform the household tasks we enjoy—perhaps cooking or gardening—as a chosen hobby. But we could pick and choose, spending less time on daily drudgery and more on chosen leisure.
That might be just as much of an improvement to human life as, say, a cheaper car or TV—and a lot more than a new app.

Unsubscribe [link removed]?