Why You Still Can’t Sell Me A Car

A millennial responds to his automobile-loving, autonomy-wary critics
Library of Congress

Last month, Popular Science ran an essay I wrote titled “Why You Can’t Sell Me A Car.” The goal was to stoke important conversations about the future of automobiles, and conversation we got—along with a flurry of hate mail.

I wrote the piece at the behest of Jake Ward, my former boss and editor-in-chief. We’d both seen several stories about automakers taking desperate measures to entice teens and 20- and early-30-somethings, also known as “millennials,” to buy cars. General Motors, for example, hired an MTV executive to tell them what might turn the tide. Ford brought on a “generational expert.”

Their brilliant solutions to boosting car sales among millennials? Knick-knacks in the dashboard, weirdly colored paints, less pushy car salespeople, and more credit cards.

As a member of the millennial generation, these and other stories irked me. Many of us grew up with landline phones, green-screened (and Internet-free) computers, CRT TVs, the original NES, and hand-written paper letters. Today a tiny slab in our pockets has essentially replaced these technologies while granting instant access to vast troves of human knowledge. We’ve witnessed these and other techy revolutions with awe, but cars haven’t kept pace. Automobiles sold today remain, at their essence, almost frozen in time. Yes, cars are safer. Yes, they’re more fuel-efficient. Yes, they’re loaded with computing devices. And yes, some are electric. But the transaction of their daily, monotonous use is the same: climb into a metal box, grip a steering wheel, maintain extreme focus, and try not to die.

New technologies woven into cars rarely solve the key problems of routine driving. When they do—especially in the area of safety—human behavior at the wheel often marginalizes the gains. Other new features function primarily as marketing ploys to squeeze consumers for more money. (Note to automakers: Americans have been working harder and earning less since before the recession.) Cars have become increasingly expensive mobile money pits that continue to mar the environment, must be driven by and among error-prone humans, and injure 2.2 million people while killing 30,000 annually. And consider the AAA’s sobering report that crashes cost society $300 billion each year.


At the risk of making myself a target for car enthusiasts, I agreed to write the essay, in which I articulated my belief that Detroit automakers can do better: by dramatically rethinking the American car. If a company that doesn’t even manufacture automobiles, for example, can coax the vehicles to safely drive themselves on bustling public roadways for more than half a million miles, then there’s no excuse for U.S. automakers not to lead the pack in disruptive technologies that could save lives, frustration, money, and the planet.

Since my essay ran, car insurance providers released two large surveys gauging American feelings toward autonomous vehicles. One was an omnibus survey of 1,000 Americans commissioned by the Chubb Group of Insurance Companies. The data suggests that 18 percent of those surveyed would buy a driverless car. A second survey by released by Carinsurance.com surveyed 2,000 Americans; 20 percent said they’d buy a driverless car. (There was no significant difference between millennials and any other age group in desire to purchase a driverless car.)

Most writers have characterized the results as a rejection of autonomous cars by consumers and lack of trust in them. I see it another way: 18 to 20 percent of licensed drivers in the U.S. is about 40 million people—and a very large market.

As for the majority opposed to autonomous vehicles? Opinion doesn’t mean they’re unsafe or a bad idea. Around the time seatbelt laws were coming into effect, for example, polls showed that most Americans rejected that idea as well. Imagine if a dealership today tried to sell cars without a seatbelt.

I actually like cars. They are marvels of engineering and permit unparalleled personal freedom in travel. But the mass audience I addressed doesn’t view cars primarily as hobbies, but as tools: for commuting to and from work, getting the kids to practice, making runs to the grocery store, and so on. I didn’t write for those who love hugging the turns of winding roads or pushing 100 mph on desolate highways or wrenching under a classic. I have no desire to pry the steering wheels from the cold, dead hands of such a passionate crowd.

Still, car-loving and car-agnostic readers alike brought up valid points, and it’s those I’d like to address to here in hopes of more vibrant conversation. Even if it means more hate mail.

Note: The following boldfaced statements are edited composites of the many reactions that readers posted online, emailed to Popular Science, and sent the author.


You don’t speak for me and I’m a millennial. I love cars just the way they are.

One of the better days of my life was driving a 1976 Porsche 930 Turbo Carrera for the first time (to the palm-sweating terror of my future wife in the passenger seat). It was a beautiful machine that a relative had spent years restoring. I never questioned why he loved the car because I didn’t have to; I understood.

Whatever feelings you might have toward a high-performance vintage vehicle, we can probably agree to dislike one thing about cars: the grind of routine travel. Daily commutes are boring, take time, cost money, and can kill us.

There was a time when Popular Science’s automotive fluff pieces were about flying cars. It’s kind of sad to see that navel-gazing millennials won’t even carry that torch.

Flying cars descended out of popular imagination and landed a while ago. Sure, they’re still horrendously expensive and haven’t yet fulfilled the promise of closing the gap between personal ground and air transportation, but they’ve indeed been around for years. One even earned regulatory clearances in 2011 for both the road and the sky.

As for fluff? If you think transportation technologies that stand to dramatically transform the modern world are fluffy, then Popular Science will continue to carry its fluffy torch.

So you want to be chauffeured around. Like many in your generation, you want to pass off your duties/responsibilities to someone or something else. (That’s called dependence.) Oh, poor you—driving is too hard and too dangerous. When history is written, your generation will be known as the “entitlement generation.”

Every U.S. generation in recent history hears similar complaints about its entitlement. It’s a cranky, worn-out, get-off-my-lawn meme that is, frankly, meaningless. So let’s retire it and actually have a conversation.

Suppose you need to send a simple message to a few dozen friends and family. Would you use a pencil, paper, envelopes, and stamps for each message—or just fire off one email and reach everyone at once? I’d prefer to have both options, but I’d use the latter on most occasions—thus saving myself the time and trouble of the former. Another question: Would you prefer to ride in an airplane equipped with autopilot that can land itself at the nearest runway during an emergency? Or a classic one that can’t if a pilot passes out in the cockpit?

The point is that we all take numerous technological advancements for granted at every turn. It’s not that big of a stretch or entitlement to desire the option of autonomous driving when it’s now tantalizingly possible and could presumably save countless lives.

Speaking of which, driving isn’t too hard for me—but it certainly is dangerous. I don’t want to accidentally hurt or kill anyone, let alone damage another’s property, and I don’t want anyone to break my stuff and/or maim me. Humans did not evolve to operate complex machinery at high speeds. Artificial intelligence can now command a vehicle more safely than trained professional drivers. If wanting such a thing as an option makes me entitled, then I’ll proudly wear that badge.

Gas wasn’t a dollar a gallon when you started driving. What alternate universe do you live in?

Actually, it was. Below is a chart of the average consumer price index for unleaded gasoline for at the time I started driving. Also note that I grew up in a Midwestern suburb flanked by cornfields and horse farms (where gas is typically cheaper than in cities).

What you’ve described is a train/bus/cab/[insert public transportation option].

Yes and no.

Trains have fixed routes, offer little privacy, and operate on their own schedules—not our whims. Buses are similar: fixed routes, fixed schedules, no privacy. Cabs? Limited service radius, expensive, no privacy, and often harbor strange smells.

Autonomous cars might traverse almost any road, yet with as much flexibility and privacy and ownership options as today’s vehicles. One would effectively operate as a private train with nearly unlimited destination and schedule possibilities. And presumably you’ll be able to change your mind mid-drive by taking the wheel or calling out a new destination.

This kind of independence is a tenet of U.S. car culture, and it’s probably why “automated highway systems” and the other tightly controlled driverless car schemes of yesteryear never took off. I believe there’s a bright future for commuters and car nuts alike—one where everyone can have their cake and eat it, too. (Note, however, that it will probably be much easier for cops to catch hoons speeding in a sea of well-behaved autos.)

You want all of us to adhere to your green philosophy. You don’t want to drive, and you don’t want the rest of us to drive either—but some of us enjoy it.

Yes to the green philosophy, which is based on decades of back-breaking scientific work. If you have a viable alternative strategy to building and organizing knowledge, and basing your positions off of that, I’m all ears. Otherwise what’s left for our children, and their children, and so on, may not be much.

More than a few people rebuked my “attack” on driving (I never launched one), stated that I hate cars (I don’t), and told me to commit suicide/shove leaky car batteries up my ass/etc. (please seek mental health counseling, angry YouTube users). It seems these critics only saw a video monologue—an off-the-cuff summary of my already condensed essay—and raised hell over that.

Alas. You can drive all you want. Keep your mint ’69 Camaro. Or buy a new internal-combusion car if driverless electrics bore you. My point is that it’s time for average, daily-commute consumers to have the option of purchasing a car that has a semi- or fully autonomous driving mode and affects the environment as little as possible. We shouldn’t be stuck choosing only models that exacerbate already deep climatological and other environmental problems that we and our forebears have created.

Dave Mosher will buy a car when and only when the car drives itself, is not made of metal or plastic, is electric and recharges instantly from a completely rebuilt countrywide power grid, and costs 10 cents to purchase. Good thing he doesn’t have any unrealistic expectations or anything.

My dig at materials in the essay was about sustainability, sourcing, and overall environmental impact. As critics have pointed out, the good news is that autos are some of the most-recycled objects in the world, with about 95 percent ending up in salvage yards. Once useful parts are removed and sold the rest is typically crushed in a compactor, shredded, and recycled. Even spare change stuck to the floor with gum can find a home in future products.

Unfortunately, about a quarter of a car’s mass still goes into landfills. Most of that material is plastic. Fortunately, sourcing for plastic these days is improving, new recycling technologies are maturing, and more biodegradable formulations are reaching larger and larger markets.

Instant charging: Getting there almost certainly requires ditching slow-charging, prone-to-explode, questionably sourced chemical batteries. The potential replacements? So-called “super” or “ultra” capacitors, which technologists promise us will charge nearly instantly, have tremendous storage capacity, and won’t degrade over time.

Sure, supercapacitors almost always seem right over the horizon (“in five to 10 years” is a common line), but one reason I’m feeling more optimistic these days is ongoing research into graphene, which won its discoverer a Nobel prize for good reason. Graphene is an atom-thick honeycomb-like carbon lattice that is very conductive and capable of storing tremendous amounts of energy. It was once notoriously difficult to manufacture, but now you can make some with pencils and tape or a CD label printer.

My own power supply dreams aside, the U.S. electrical grid already needs an extensive overhaul. It wasn’t designed to handle modern uses, loads, or catastrophes. Small disruptions can and often do cascade into epic failures. And may Zeus help us if a truly significant disruption, such as a powerful solar storm, blasts the Earth. Why not solve multiple significant problems in one go? Car manufacturers could put serious force behind the push to secure energy sources for their next-generation vehicles while easing big environmental, supply, and reliability problems.

The U.S. must start thinking big and thinking differently about how, where, and when to get renewable energy, and by what means to transport and store it. It’s probably unnecessary to erect eye-soring forests of wind farms everywhere; we could pipe in electricity generated by solar collectors or wave generators from halfway across the country. With the right resources, it may even be possible to beam it down from space. The future is wide open for innovation and implementation.


Costs: Any truly disruptive technology is horrendously expensive when it first reaches the market. Then, inevitably, the price goes down—often way, way down. The first cell phones, for example, cost thousands of dollars. Now you can get a prepaid burner phone for less than $20. We should expect self-driving cars to follow a similar trend but I admit that, for now, the complexity and cost of these systems is nothing to sneeze at. Consider, for example, the $80,000-or-so cost of the LIDAR system that Google uses—ouch.

Mass production and consumer demand will help reduce costs, plus whatever unknown advancements occur between now and when Detroit sells its first semi-autonomous model (a Toyota representative told me recently that he expects semi-autonomy to be common in 10 years and full autonomy within 20). Another thing to consider is long-term savings on insurance, deductibles, maintenance, and fuel. Then you weigh the benefits of lower stress, increased work productivity, and—dare I say it?—more sleep.

As for my unrealistic expectations, I’ll say it multiple times: I never said it would be easy. My characterization of “that’s all you have to do” was meant to be tongue-in-cheek. It’s going to take extremely hard work in the form of expensive R&D, strong incentives/subsidies, overhaul of federal and state laws, favorable court decisions and legal reviews, and pervasive consumer education, among other things. Automakers are sitting on a lot of money and influence that can help bang the drums—and ultimately to their benefit as well as that of the rest of the world.

To build your rosy world of the future, auto manufacturers will have to take on incredible research and development costs. That’s a foolhardy demand at best.

The biggest automakers already spend billions of dollars on R&D, consistently placing them among the top 20 companies that invest in research annually. Most of them are dipping into these war chests to develop at least semi-autonomous driving systems.

So why am I bent out of shape? First, those budgets aren’t that big, relatively speaking; they usually comproise single-digit percentages of an automaker’s revenue. Second, the recent trend for U.S. automakers is to spend less on R&D (“don’t eat your children,” as the saying goes). Third, and perhaps the most important—yet the most fuzzy—is how that R&D money gets spent.

I’d wager the amount appropriated to developing full autonomy is small, with most R&D dollars going toward legacy technologies and programs—not reimagining the future of the private passenger automobile.

I never said it’d be easy. I recognize that developing autonomous vehicles is extraordinarily difficult work that’s fraught with obstacles (physical ones notwithstanding). But I’m rooting for U.S. auto manufacturers. I think they have the financial and intellectual capital to get it done, and I’m not alone in my optimism about the future. Just look at what Daimler AG (via Mercedes-Benz) can do today.

If U.S. automakers can’t find money to spend on full autonomy, here’s one idea: Divert a fraction of their multi-billion-dollar advertising and marketing budgets. I enjoy Will Ferrell’s humor as much as the next person, but do we really need Ron Burgundy to shill us cars? I’d rather see tangible innovation in autonomy and energy storage, and faster.

Get rid of nuclear and fossil fuels and rely solely only on renewables? Good luck. That’ll cost hundreds of billions worth of investment over decades, and car manufacturers can’t get it done. A better idea is to encourage the growth of natural gas and buy time to develop next-generation energy technologies.

The video gives the impression that only renewable energy will satisfy me—and that’s not the case. It may seem noble to end the use of fossil fuels entirely, but it’s not possible, practical, or advisable right now.

As we reported in the June 2013 issue of Popular Science, the path to American energy independence (and a greener world) is at least a few decades away. The situation is complicated, as with most things in life. We’ll need to develop a mixture of solar, ocean, wind, energy from waste, and, yes, fossil fuel technologies to get there.

A key part of the energy picture—especially in discussing driverless electric cars—is to highlight is the enormous demand by transportation for petroleum. In fact, moving things around with internal combustion engines takes up nearly a quarter of the entire U.S. energy budget.

Automakers have serious clout among oil companies and the government, and that will only be more true if they begin abandoning ICEs. They could push for vast improvements to the grid, both in basic infrastructure and sources of renewable energy. It may sound strange now, but car companies could eventually deal major blows to the problems of fossil fuel dependence and climate change.


Nuclear: I’m not advocating we give it the boot. However, our reliance on dated pressurized water reactor technologies needs to end. The constant generation of waste that takes millions of years to become safe is one issue. Fukushima and other meltdowns have clearly demonstrated the other, which is that these kinds of reactors are not walk-away safe.

There are almost certainly better ways to extract energy from nuclear materials. Thorium molten salt reactors, for example, promise to chew through yesterday’s nuclear waste, turn the radioactive chaff of rare-earth oxide mines into fuel, and safely shut down on their own (even if—for some strange reason—all humans instantly vanished from Earth). They might even ease or end NASA’s planetary exploration woes by generating plutonium-238.

Unfortunately, the redevelopment and refinement of the technology could cost billions of dollars, and the economics of the thorium supply remain untested.

Natural gas, Obama’s bridge-to-the-future energy source, was a nice idea but it is turning into a nightmare. Aside from the known problems with fracking—tainted water sources, financial and property abuses, and earthquakes, to name a few—there is a much larger threat posed by extracting gas from the ground: rampant leaks. Unless the holes in ancient pipes for delivering gas are plugged across the U.S., the methane-laced gas leaking from pipes stands to accelerate climate change more than any other fossil fuel used today.

You wrote that automakers got the government to build highways, but that’s not true; war and commerce did.

Generalizing is too often a necessary evil in magazine writing, but I never discounted the roles of war (rapidly mobilizing military resources) or economic progress (rapidly moving products and materials) in getting highways built. Rather, I chose to point out the most powerful player in the tight space I was given.

Yes, German highways made a big impression on Eisenhower during World War II. Yes, the Cold War’s nuclear threats increased the desire to have some means to quickly evacuate cities and move around the chess pieces of war. And yes, rural regions wanted access to the commerces of big cities via extensive roadways.

Automakers, however, were instrumental in uniting these interests and making highways happen through decades of lobbying (and the undermining of public transportation systems). It’s not hard to understand their motivation. More roads meant more cars, more customers, and more money.

If autonomous cars become the norm, no one will know how to drive if a system fails.

Most people can’t drive already, so what’s the problem?

Jokes aside, this is a valid concern—that is, if lawmakers abandon rigorous tests for driver’s licenses. Even if self-driving cars become the norm, there will always be a human desire to grab the wheel and override the docile nature of an autonomous vehicle. I suspect that pressure from automakers and motorheads alike will ensure that manual driving is a standard feature and prevent licenses from vanishing.

Driverless cars will never happen because the liability risks to manufacturers will be too great. They will go bankrupt being held accountable for any accidents.

People have dreamt up all kinds of trolley problem scenarios for driverless cars as a means of (mentally) exploring their safety features and ethics. I’ll make one up now as an example: An autonomous car is driving 55 mph down a two-lane highway. A group of five children leap into the car’s lane while a single kid steps into the oncoming traffic lane at the same time. There isn’t enough distance to brake to a full stop. What will the car do? a) Swerve, brake, and hit the one kid; b) brake hard and hit the five children; c) steer into a ditch, possibly killing the driver and passengers; or d) hand control back to the driver, allowing him or her to choose a fate?

My guess is as good as anyone’s, but others surmise that the best answer to this scenario is another question: Who gets sued?

Automakers’ billions, compared to an individual’s insurance policy, makes the answer seem fairly clear. Yet the legal precedent for driverless cars is essentially nonexistent. This may partly explain U.S. automakers’ hesitance to plow forward with such technologies, i.e. nobody wants their car to be the first to wreck in an increasingly litigious nation. After all, the U.S. spends more than 2 percent of the GDP on tort litigation annually.

On the upside, there’s room for intense discussion and—for the boldest and brightest among us—an opportunity to blaze a responsible path forward.

I’ll give Patrick Lin the final word on this, since his ethics-of-autonomous-cars piece at TheAtlantic.com captures this well: