I remember back in high school having a discussion about the risks of the Internet. It was the mid-2000s, a time when the last remnants of the Internet’s patina of novelty were wearing away. Facebook was getting popular and Wikipedia and Google had obviated the need for "surfing the web" for information - they just brought it right to you. In our discussion, someone made a point about how she was worried that the Internet could be used to track down people in real life. She told a hypothetical story about how much someone could learn about you just from knowing your name. First, she said, they'd find your Facebook profile, then use landmarks in your public pictures to identify what town you were in. Then, her story went, they'd follow tiny Internet clues like Sherlock Holmes to continue to narrow down your location. Somehow, and I don't remember the details, but I remember it being really convoluted, they would, just from your name, figure out your address. And that was the end for you because once strangers from the Internet could find your address, you were basically as good as dead.
Does anyone remember what a phone book was? For those who don't, it's a list of every adult's name, address, and phone number in an area. They were sent to your house for free. I would use them all the time to find people's numbers or addresses; it was a totally normal thing to do. Finding someone's address from their name wasn't only possible, it was the entire point of the book. But instead of protesting against these books, my classmate was adamantly protesting against Facebook because of its potential as an incredibly inefficient phone book.
Why is this? I think it’s because by that time, phone books were considered normal - they were part of the status quo. They had been around a long time and, obviously, people weren't being chopped up right and left, so they must be OK. But Facebook was new, so who knew what could happen?
I think there's an incredibly strong bias that people have for the status quo, which I call the status quo bias. The status quo bias is an implicit belief that the current state is preferred over alternative states. Despite it being such a strong bias, I don't hear it discussed much. I think that's a mistake, and in the post, I'll walk through some more examples illustrating it.
Let's take a simple example. Why don't people wear their seatbelts on buses? At the risk of sounding trite, I think the true answer is “because they don't,” or, at least, “because everyone around them doesn’t.” If you ask someone, you might hear some claim about the number of accidents that buses are in, but is this the real reason? If advancements in car safety technology dropped the accident levels to whatever it is today with buses, would people stop wearing seat belts? No, that would be considered crazy. One might counter that buses go slowly mostly in urban settings, but, again, I think this is specious. I've been on plenty of long-distance, highway buses and people still don't wear seatbelts. Or you might hear something about how because the seat in front of you is closer it's less dangerous. No one does these measurements and uses the same logic in the backseat of a car. People don't wear seatbelts on buses simply because that's what they're used to. It's the status quo bias.
OK, seatbelts on a bus might not seem like a big deal, but let's take that same bus and talk about who's driving it. Currently, the answer is that one person is driving it. But at some point, that job will be automated and there won't be anyone required to drive that bus. This is certainly causing a lot of discussion about whether this kind of automation should be allowed. There are an endless number of articles about how robots are threatening our jobs and journalists telling us that we should "be very afraid".
It's natural to prefer the status quo. After all, you know the current state of affairs and the consequences of a new paradigm are always hard to predict. With any new technology, it's important to carefully consider how it's deployed. These aren't just jobs in some economist's spreadsheet, there are real people with real lives working these jobs. And when these jobs are no longer needed, these people should be treated with respect and assisted in the transition, whether that's through job training, early retirement, or something else. We need to simultaneously consider the human experience while not letting the status quo bias (or other biases) blind us to what's going on. Once you separate these things in your mind, it focuses you on two things:
We should work hard to ensure that people whose jobs are replaced are treated well, and
There is no reason to hold onto obsolete jobs.
If, after self-driving buses become equally safe, you're still advocating for keeping those jobs, let me ask you this: "Why aren't you advocating for two bus drivers per bus right now?" If when the number of required drivers goes from 1 to 0 and you're still advocating for 1 driver, how's that different than advocating for 2 right now? There is nothing inherently special about the ratio of 1 driver to 1 bus. Except that, of course, it's the current status quo.
Some jobs have multiple purposes and only a portion of that is automated. Perhaps, people like it when a friendly bus driver greets their children on the way to school. That's great, let's optimize for that. Instead of being a bus driver, we could make the job more of a friendly chaperone who can focus even more on being with the children.
Just like no one advocates for doubling the number of bus drivers, advocating to bring back old jobs is almost unheard of. People talk about protecting all the jobs that will be lost to AI, but no one talks about all those jobs that were lost due to spreadsheet software. This is not a joke. Think about all the number crunchers that lost their jobs. For more, check out Planet Money’s excellent podcast on the impact of spreadsheet software.
Again, there are people in these jobs, and we should be deliberate in assisting their transition. But, putting that aside, once you acknowledge that the job is no longer needed, then what's the difference between fighting to keep it and fighting to bring back an old job? Should we bring back elevator operators, switchboard operators (working on switchboards that aren't connected to anything), and pinsetters at bowling alleys? The only difference is the status quo.
Status quo bias is bad for two reasons. It causes people to resist change too much and question the current status not enough. The first we’ve talked about, but the second can be just as harmful. It prevents people from questioning the current state of affairs, and sometimes that state of affairs is in need of correction.
We're a nation that's been facing an obesity problem for decades. Children getting diagnosed with type 2 diabetes was once unheard of but now is common. But still, when I wait in the checkout aisle, I'm surrounded by rows of candy bars. Is this a good idea for a nation dealing with a massive obesity problem? I think it's a horrible idea, but I've never heard it discussed. Where are the advocacy campaigns to get stores to stop doing this? It's not even in the national conversation.
Here's another one. Some newspapers try to scare us about AI by saying things like "AI combined with autonomous weapons could launch an era of indiscriminate killing the likes of which civilization has never seen before." But here's the thing that they all seem to miss. The most dangerous weapons are those that already exist and were built decades ago. Intercontinental ballistic missiles delivering nuclear payloads can kill hundreds of thousands of people anywhere in the world in 15 minutes. That's what's scary. I've written about this in the past so won't go into it more here.
So let's recognize this as a bias, and challenge ourselves when we feel ourselves falling victim to it.