A few things happened when SpaceX’s uncrewed experimental rocket blasted off and then exploded mid-flight last week. The engineers who’d designed it let out a deep sigh, maybe a couple of groans. The Federal Aviation Administration, which regulates rocket launches, began a standard investigation into what happened. Elon Musk congratulated his staff on a good start. And in a small nearby city, ashlike debris rained down from the sky, covering everything in a layer of grime.

The giant Starship rocket lifted off with so much force that it not only blew a massive hole in the launchpad’s foundation but also kicked up a cloud of sand and soil that reached Port Isabel, located about six miles northwest of the company’s launch facility in Texas. The launch also shook houses and shattered at least one window, The New York Times reported, though there were no reports of injuries or damage to public property in the aftermath.

Residents knew this launch was happening. They are well aware that they share a neighborhood with a powerful company intent on creating a spaceship that can fly to both the moon and Mars. And they knew it would be loud; SpaceX had warned Cameron County residents of that. Still, no one had warned residents about the dusty drizzle. (SpaceX did not respond to requests for comment.)

[Read: Elon Musk is maybe, actually, strangely going to do this Mars thing]

SpaceX’s presence in southern Texas has received mixed reviews over the years, particularly from people who live even closer to it—just down the road from the ever-growing launch facility—than Port Isabel. SpaceX’s takeover of Boca Chica Beach, a quiet coastal paradise along the Gulf of Mexico, is a nearly decade-long saga of Musk and his enterprise getting their way, sometimes to locals’ detriment. And that was before SpaceX attempted its biggest launch yet, trying to hurl Starship into orbit. Starship was the most powerful rocket ever built, and more prototypes will follow—SpaceX will attempt as many launches as necessary to prove that its design actually works. As Musk expands his beachfront space city and its potentially fiery attractions, his company risks becoming an even more unwelcome neighbor.

SpaceX began construction of its launch site on Boca Chica Beach nearly a decade ago, bringing in so many truckloads of materials that the only highway leading to the area started cracking. As the spaceport took shape, SpaceX bought out some of the residents of nearby Boca Chica Village. The company rebranded the area as Starbase, and SpaceXers, including Musk himself, moved in, replacing the retirees who felt forced to leave. SpaceX instructs Boca Chica residents to evacuate and closes off the beach during launch activities, including last week’s test—the first of its kind and the most destructive yet.

The earliest Starship prototypes were small, pudgy things, topping out at 65 feet. The latest prototype—a sleek spaceship stacked on top of a 33-engine rocket booster—was a nearly 400-foot behemoth, taller than the Statue of Liberty and more powerful than the rocket that propelled the Apollo astronauts to the moon. All of that power had to go somewhere, and much of it went into breaking up the launchpad and scattering debris for miles. A reporter with The Brownsville Herald came across a huge chunk of concrete on the sands of Boca Chica Beach a few days after the launch. Fish and wildlife authorities said yesterday that debris was scattered across 385 acres of SpaceX-owned and state-owned parkland, Bloomberg reported, and even sparked a small fire in a nearby state park. There were no reports of dead birds or other wildlife, they said.

[Read: Why SpaceX wants a tiny Texas neighborhood so badly]

SpaceX had predicted that the debris from an explosion would be contained within a 700-acre area (about one square mile), but the Texas division of the U.S. Fish and Wildlife Service reported that a “plume cloud of pulverized concrete” dropped particles miles beyond that zone, all the way to Port Isabel. City officials said the dust was not harmful to human health; local environmental groups, which have criticized SpaceX’s operations for years, say they’re evaluating the dust’s potential effects on people and wildlife.

SpaceX seems to have also miscalculated how the launchpad would fare beneath a vehicle such as this one. Starship lifted off without the special trench system found at major launch sites, which helps dampen the impact of launch. Workers had begun building one but hadn’t finished in time. Musk later tweeted that the company “wrongly thought” the launchpad would be unscathed. Yes, the Starship program is an experimental one, and SpaceX doesn’t shy away from potential explosions during testing. But the decision to launch with little more than a slab of concrete in the ground doesn’t exactly inspire confidence in the company’s approach to safety.

A week out from Starship’s dramatic, short-lived flight, Starbase is as busy as ever, with technicians preparing for repairs. Musk says they’ll be ready to try again in a month or two, but any Muskian timelines must be taken with a big grain of launchpad dust. Engineers will spend the coming weeks poring over data from the test, learning for the next go. Some locals would probably prefer that they also learn how the surrounding areas might be affected if another Starship starts tumbling in the sky and breaks apart. After all, a rocket with 33 engines, as Musk himself said earlier this month, is like “a box of grenades.” Someday, if SpaceX succeeds in launching rockets on Mars, it may really be able to operate without concern for its nonexistent neighbors. It’ll have entire craters to itself. But the starting point of that future is here, on Earth, and SpaceX can’t avoid sharing it.

The real problem with space is there’s not enough of it—at least not when it comes to places to put people. With billionaires funding their own space programs and investors pouring cash into new start-ups aimed at building a true space economy, millions of humans may end up working and living away from Earth in a century or two. If that happens, all of those people will need somewhere to live. But as of yet, no would-be captain of space industry has proposed a viable housing plan.

Mars, which gets a lot of attention as ground zero for humanity’s future, is really just a good place to die. The Red Planet is a frozen desert with a thin atmosphere and no protecting magnetic field. The threat from high-energy solar radiation is so extreme that cities would have to be built underground or covered by thick domes. And the surface of Mars (or any planet or moon) sits at the bottom of a deep gravitational well (that is, a high-gravity region) that rockets must climb out of or carefully drop down into. This makes visits between the surface and space an expensive proposition: It takes a lot of rocket fuel to climb up and down those wells, and rocket fuel doesn’t grow on trees (and trees don’t grow on Mars).

Any serious efforts to live outside Earth will have to go beyond the immediately obvious destinations such as Mars. There are options that don’t involve the great expense of climbing out of gravity wells and the great danger of falling back down them again. Based on recent work, my colleagues and I may have found a candidate with none of Mars’s tricky problems. If we’re right, our real future in space may be not on the surface of planets, but inside asteroids.

NASA has been considering non-planet, non-moon options for our future space housing as far back as the early 1970s, when it began exploring designs for “space habitats.” Then, in 1976, the Princeton physicist Gerard K. O’Neill published The High Frontier: Human Colonies in Space. The book, which became an instant classic, was full of plans for permanent space cities contained in shining metal cylinders, the largest of which was 20 miles long and four miles across.

[From the January/February 2015 issue: 5,200 days in space]

These “O’Neill cylinders” addressed the biggest and most fundamental problem of not living on a planet: no gravity. Research indicates that when people live in zero gravity for more than a few months, their eyeballs bulge out, their retinas can detach, their muscles atrophy, and their bones become brittle. While there aren’t many data on the problem, evidence suggests that humans may require somewhere around one-third of Earth’s gravity in order to function properly. That’s one reason people such as Elon Musk are so bullish on Mars: Its surface gravity is just above that one-third limit. An O’Neill cylinder wouldn’t be massive enough to generate significant gravity, but O’Neill designed them to rotate around their long axis, producing gravitylike centrifugal force. Residents would live happily on the cylinder’s inner surface, pulled “downward” away from the center.

The stunning artwork that accompanied O’Neill’s designs showed futuristic communities set amid beautiful, landscaped parklands where the horizon curved upward. The pictures fired the imagination of a generation of space nerds, including me and, more notably, Jeff Bezos, who owns the rocket company Blue Origin and is a big fan of space habitats. But O’Neill cylinders have a crucial problem: Bringing millions of tons of raw materials up from a gravity well (even a shallow one such as the moon’s) and then fabricating them into the necessary steel beams, trusses, and construction essentials would be prohibitively expensive. I’ve seen an estimate of $100 trillion a cylinder.

One could conceivably get the material from asteroids, which don’t have significant gravity. But even after grinding the asteroids down, you’d still need giant space factories for transforming the extracted raw materials into O’Neill cylinders. That wouldn’t be cheap either. The one thing asteroids have going for them is their sheer quantity: Thousands of the flying space mountains zoom around on orbits that pass near Earth. Actually, if you think about it, that’s a lot of potential real estate.

[Read: NASA is practicing asteroid deflection. You know, just in case.]

To be clear, we can’t live on asteroids (too little gravity, too much radiation). And we can’t live in metal tubes made from asteroids (too expensive). But there may be a third option: living inside asteroids.

It’s an idea that science-fiction writers, at least, have taken notice of for more than a decade. Hollowed-out space rocks, set into rotation to generate centrifugal force, play a central role in Amazon’s TV series The Expanse (based on the books series of the same name). In that fictional future, about 100 million “Belters” live in asteroid cities, where the thick rock walls offer free, natural protection from high-energy solar radiation. (The Expanse is on Amazon for a reason: Bezos loved the show so much that he rescued it from cancellation from the Syfy channel).

But science fiction is one thing; the laws of physics are another. I and my colleagues at the University of Rochester wanted to find out if asteroid habitats are really a possibility. Within the next several centuries, humans may very well have the technology to dig out big living spaces in an asteroid and set it spinning. But when my colleague Alice Quillen, an asteroid expert, and I ran the calculations on that plan, we found that spinning a large asteroid fast enough to create artificial gravity would crack and fracture the asteroid rock matrix, leading to the whole asteroid flinging apart. As a huge fan of The Expanse, I was bummed out by this result. But Alice came up with another approach.

[Read: A handful of asteroid could help decipher our entire existence]

Most asteroids, it turns out, aren’t solid rocks. The ones smaller than about 10 kilometers across are a mixture of sand, pebbles, rocks, and boulders held together by the weak force of their own gravity. Spinning such a body would send its component parts flying into space immediately—unless, as Alice pointed out, you could hold them together. All we needed was a very big bag.

Imagine a swarm of robots easing up to one of these rubble-pile asteroids and covering it with an ultra-strong, ultralight elastic webbing. (Carbon nanowires, which can be as thin as a few atoms, are a good option. They’re currently produced in only small amounts, but we could manufacture them en masse in the future.) Once the cylinder-shaped “bag” is set, the asteroid could then be slowly spun via rocket motors anchored deep in the rubble. As the rubble pile spins faster, it would begin flinging out those pebbles, rocks, and boulders, pushing the carbon-nanowire webbing outward with them.

Rendering of a space habitat made by spinning an asteroid (Michael Osadciw/University of Rochester)

At some point, the bag would hit its maximum extension, and the expanding rubble would slam into now-rigid webbing. Debris flung into the taut bag would be compacted, forming a giant, hollow, concrete cylinder. Once the dust clears, towns, cities, parks, and farmland could all be built on the inner surface of the cylinder, just like in O’Neill’s designs. That surface could be enclosed with a transparent roof to hold in an atmosphere; imagine a bicycle tire with clear plastic running around the inner circumference. Outside the living area, the thick concrete walls would protect against radiation.

[Read: Just like that, we’re making oxygen on Mars]

Led by a very smart engineering Ph.D. student named Peter Miklavcic, our team ran a series of simulations to study the plan. It seemed to work: Our models showed that a small rubble-pile asteroid, just a few football fields across, could be expanded into a cylindrical space habitat with about 22 square miles of living area. That’s roughly the size of Manhattan, where more than 1.5 million people currently live. Multiply that by the solar system’s tens of thousands of asteroids, and The Expanse’s 100 million space dwellers could be easily accommodated.

Humans do not yet have the technology to build asteroid cities. It’s also possible that we will never have the kind of industrial space infrastructure necessary to make this idea a reality. Maybe it won’t be possible to manufacture enough carbon nanowire to make the asteroid bags. Maybe social systems are unstable in permanently enclosed habitats. These are all possibilities. But the paper summarizing our results, which we published last year, suggests that the overall plan is viable. The technologies we proposed using don’t break any laws of physics; they are reasonable extrapolations of capacities we have now. Such extrapolations can seem fantastical, but they prove true more often than you might think. Remember that in 1900, no one had ever flown an airplane. Now, as you read this, hundreds of thousands of people are safely hurtling five miles above the ground at hundreds of miles an hour.

[Read: When a Mars simulation goes wrong]

Perhaps future generations will blaze ahead with Mars settlements anyway, deciding to drop down treacherous gravity wells and take their chances on an unforgiving desert. But I’d rather imagine the solar system hung with thousands of asteroid cities, each a spinning jewel of human creativity and promise.

The Republican plan for 2024 is already failing, and the party leadership can see it and knows it.

There was no secret to a more intelligent and intentional Republican plan for 2024. It would have gone like this:

1). Replace Donald Trump at the head of the ticket with somebody less obnoxious and impulsive.

2). Capitalize on inflation and other economic troubles.

3). Offer plausible ideas on drugs, crime, and border enforcement.

4). Reassure women worried about the post-Roe future.

5). Don’t be too obvious about suppressing Democratic votes, because really blatant voter suppression will provoke and mobilize Democrats to vote, not discourage them.

Unfortunately for them, Republicans have turned every element of the plan upside down and inside out. Despite lavish anti-Trump donations by big-money Republicans, Trump is cruising to easy renomination. Rather than capitalize on existing economic troubles, Republicans have started a debt-ceiling fight that will cast them as the cause of America’s economic troubles. Worse for them, the troubles are fast receding. Inflation is vexing, but the recession that Republicans hoped for did not materialize: Instead, Joe Biden has presided over the fastest and steepest unemployment reduction in U.S. economic history since he took office in January 2021.

[David Frum: Biden laid the trap. Trump walked into it.]

The big new Republican idea to halt the flow of drugs is to bomb or invade Mexico. Instead of reassuring women, Republican state legislators and Republican judges are signaling that they will support a national abortion ban if their party wins in 2024—and are already building the apparatus of surveillance and control of women necessary to make such a ban effective. Republican state-level voter-suppression schemes have been noisy and alarming when the GOP plan called for them to be subtle and technical.

It’s early in the election cycle, of course, but not too early to wonder: Are we watching a Republican electoral disaster in the making?

Biden’s poll numbers are only so-so. But a presidential election offers a stark and binary choice: this or that? Biden may fall short of some voters’ imagined ideal of a president, but in 2024, voters won’t be comparing the Democrat with that ideal. They will be comparing him with the Republican alternative.

An American must be at least 36 years old to have participated in an election in which the Republican candidate for president won the most votes. An American must be at least 52 years old to have participated in two presidential elections in which the Republican nominee got the most votes.

Despite this, over the past 30 years, the GOP has succeeded in leveraging its smaller share of the vote into a larger share of national power. That same 36-year-old American has lived half of his or her adult life under a Republican-controlled Senate, and even more of it under a Republican-majority House of Representatives. Through almost all of that American’s adult life, Republicans have held more than half of all state legislatures. Conservative dominance of the federal courts has become ever more total in the past two decades, culminating in the Supreme Court’s reversal of Roe v. Wade.

[Tom Nichols: Trump did it again]

Some of the Republicans’ leverage can be explained by the American electoral system’s tilt against metropolitan areas. Some of their success is due to luck. The GOP’s big year of 2010 also happened to be a redistricting year, so one successful election translated into a decade of more comprehensively gerrymandered state legislatures. (Democrats have not had a big win in a redistricting year since 1930.)

But the tilt is not infinite, and the party’s luck is running out. Republicans have suffered a series of heavy defeats since the rise of Trump: loss of the House in 2018, loss of the presidency in 2020, loss of the Senate in 2021, losses at the state level in 2022 (Democrats won net two governorships and net four legislative chambers).

Trump-era Republicans have difficulty absorbing and reacting to negative news. Led by Trump himself, they misrepresented 2016 as—in the words of his former adviser Kellyanne Conway—a blowout, historic landslide. They misrepresented 2020 as an election that they deservedly won, but that was stolen from them by fraud and chicanery. Out-of-office Republicans like Paul Ryan will acknowledge on CNN that Trump lost. But they won’t say it on Fox News. Trump’s own leading party rival, Florida Governor Ron DeSantis, won’t say it. And if Trump is indeed the primary winner that he insists he is, what on Earth is the case for denying this political superstar the third nomination he wants?

[David Frum: Is Ron DeSantis flaming out already?]

The Democrats, by contrast, are a party that has trouble absorbing and reacting to good news. Few Democrats predicted that the party would do as well as it did in 2022. Most feel deep dread and anxiety about 2024.

Maybe it’s good to guard against complacency. The American electoral system’s tilt against Democratic-voting regions remains as pronounced as ever. The Senate map is especially unpromising for Democrats. Yet it’s also important to understand that although America is intensely and bitterly polarized, it is not evenly polarized.

The potential strength of the Democratic coalition is greater than that of the Trump coalition. The Democratic disadvantage is that their coalition spans a lot of groups that face extra difficulties casting a ballot: renters, college students, hourly workers, single parents, people who don’t own cars. The American voting system has been engineered to deter and discourage them.

If motivated to turn out, however, those deterred and discouraged blocs can swing elections. In 2018, 36 percent of 18-to-29-year-olds turned out, the highest level recorded. Their votes helped change control of the House. Turnout of this cohort in 2022 finished second only to what it had been in 2018, and those votes altered the political complexion of many state legislatures. The state that had the highest youth turnout in 2022 was Michigan—not so coincidentally, the state where Democrats scored some of their biggest gains, flipping both chambers of the state legislature from red to blue.

Chief among what motivates voters who face obstacles is hope. People will endure and overcome barriers when they feel that their vote can make a difference. If Democrats succeed in communicating hope in 2024 that young people can contribute to a decisive defeat of Trump and MAGA extremism, then that is what they will do.

[Peter Wehner: The institutional arsonist turns on his own party]

This cycle, that hope is well founded. Republicans are doing everything wrong. They are talking to their voters about Trump’s personal grievances and about boutique culture-war issues that their own base does not much care about, such as the state of Florida’s “war on Disney.” At the same time, Republican leaders are confronting Democratic voters with extremist threats on issues they care intensely about: bans on abortion medication by mail, restrictions on the freedom of young women to travel across state lines, attacks on student voting rights, proposed big cuts to Medicaid and food stamps in the GOP debt-ceiling ransom demand. Republicans offer no economic message and no affirming vision, even as they make new moves to police women’s bodies and start a land war in Mexico. They are well on their way to earning a deep, nasty defeat—and the smell of that defeat may be an additional draw to the polls for the Democratic-leaning constituencies that will inflict it.

Of all the major-party candidates to run for president since 2000, only one scored worse than Trump in the popular vote: John McCain in 2008. That was not a personal verdict on McCain. He was running for a third Republican term in the throes of the worst economic catastrophe since the Great Depression and against the backdrop of the most grinding military frustration since Vietnam.

Biden’s reelection-announcement video, released yesterday, defines the principal issue at stake in 2024 as “freedom.” From the New Deal to Trump, “freedom” was a Republican slogan; “security” was its Democratic counterpart. But Trump, together with DeSantis, has completely rebranded the GOP as the party of bossing around women, minorities, and young people.

If Trump secures the GOP nomination to run for a second term in 2024, the conditions are all in place to transfer the title of “worst popular-vote loser of the century” from the great Arizona senator to the putsch-plotting ex-president. Trump’s own party is doing its part to deliver this debacle. Soon enough, all Americans will have the opportunity to do theirs.

A good way to learn psychology is to ruin two birthday parties.

Take it from me: I’ve got a Ph.D. in experimental psychology, but I didn’t really understand my own field until I started showing up at strangers’ birthday parties because science told me to. And now that I’ve inadvertently wrecked multiple get-togethers, I finally know the true meaning of psychology.

It happened in Atlanta, just after the 2018 meeting of the Society for Personality and Social Psychology. The place was buzzing with a new idea: Talking to strangers, research showed, could be surprisingly delightful. A few years earlier, two researchers had persuaded a bunch of Chicago commuters to talk to one another during their train ride in exchange for a free banana; the participants reported having an unexpectedly wonderful time. Now several other teams were reporting similar results. A few of my friends had discovered that strangers tend to like each other a lot when they first meet, but underestimate how much the other person likes them. I was at the conference hawking my own, related findings: Conversations rarely end when people want them to, but people have a nice time anyway. (Two of my participants flirtatiously exchanged numbers after meeting each other in my study, so, if nothing else, I was running an extremely inefficient dating service.)

So when I ended up with an extra day in Atlanta, I thought, Why not listen to all these studies and go meet some strangers? And I knew just how to do it.

I’m an escape-room enthusiast––that is, I like to pay people about $35 to lock me in a converted office space with a bunch of friends and some themed puzzles. I have done this more than 140 times, turning what might otherwise be a silly hobby into the most annoying aspect of my personality. Many escape-room websites allow you to see how many people have already signed up for a certain time slot and, as if predicting my pro-stranger epiphany, they even allow you to join those preexisting groups without consent. I was so excited that I signed up for two nearly full rooms. Unforeseen pleasures, I thought, here I come!

I began to realize my mistake on the way to the first venue. I was running late, so the woman who ran the room called me to check in. “When are you arriving?” she asked, but something in her voice said “Why are you arriving?” I wondered, Was I doing something weird? When I got there, the answer was immediately, obviously, yes. The rest of the team was waiting, and they were clearly unhappy to see me, but they were being heroically polite about it. They were six 20- and 30-somethings, profoundly normal, not the kind of people who would get hopped up on psychology studies and go crash someone else’s escape room. In another life, we might have been friends. But we weren’t. They were friends; I was a stranger. I got the impression one of them was having a birthday, but I wasn’t sure, because it’s weird to tell strangers that it’s your birthday unless you’re at a restaurant and angling for a free cupcake. The suspected birthday boy looked saddest of all to see me.

“You must be Adam,” he said.

I figured I could win over my teammates by being really good at escape rooms. All I had to do was solve a big puzzle; we’d all cheer, and then we’d get to talking about our dreams of starting our own escape room (the theme would be “friendship”), and then maybe we’d go throw a Frisbee around and we’d exchange numbers and become lifelong pals, just like the people in the psychology studies did.

“Maybe this does something,” I said, holding up a Ping-Pong ball.

“What does it do?” one of the guys asked eagerly.

“I don’t know,” I said.

I had forgotten that a key part of doing escape rooms is communication, and talking to people is hard when they all know one another and you don’t know them, and especially when you appear to be ruining their birthday party. In fact, no one really talked to me for the rest of the room, except when I said things like “There’s something about this clock” and they said “Uh-huh.” They did let me open one of the combination locks, even though I wasn’t the one who figured out the code, which was generous of them.

As soon as we got out, I skedaddled. I wanted to leave these poor strangers as soon as possible. But I also had a second room to get to––I had already paid for it, and as every social scientist knows, you’re supposed to honor sunk costs.

[Read: The surprising benefits of talking to strangers]

This one was unmistakably a birthday; the room was a prelude to a party. The gang of friends was slightly older this time, all of them new-ish parents who had left the kids with a babysitter for a little adult fun. I forget the birthday gal’s name, but I do remember that her daughter’s name was pronounced “Sir-see,” like the terrifying character from Game of Thrones who incestuously had three children with her twin brother and was later crushed by rocks.

Worst of all, the room was Prohibition-themed, and all the other guests were wearing old-timey clothes. The ladies wore flapper dresses and fascinators; the men wore suspenders and spats. I wore jeans and a plaid button-down, which was now thoroughly soaked with the kind of sweat you only produce when you know you’re doing wrong.

Once again, I embarrassed myself by solving virtually no puzzles. I walked around with fake thoughtfulness, picking things up and putting them down. “There’s probably a second room; there’s usually a second room,” I once offered, helpfully, to no response. My big win was finding a key and then, every five minutes or so, saying, “I have a key if anyone needs a key,” and when someone finally needed a key, I went, “The key!” and unlocked their padlock for them.

When we got out, the group assembled for a photo. “Let’s do one with Adam and one without,” someone offered, weakly.

“That’s okay, gotta go!” I said, already fleeing.

Where were my unforeseen pleasures? Where was my surprisingly delightful time? Where was my flirtatiously exchanged phone number? Why had science lied to me?

Psychologists sometimes act like we’re compiling a how-to book for life. Year by year, we scratch out the old wives’ tales, folk theories, and cognitive biases, and then replace them with evidence-based guidance for making better, happier decisions.

We are not compiling a how-to book for life. Many of our studies fail to replicate, but even if every paper were 100 percent true, you could not staple them together into an instruction manual, for two reasons.

First, people are just too diverse. Almost nothing we discover is going to be true for every single human. In my own research, for example, some strangers became fast friends, but others spent two painful minutes asking questions like “So, uh, do you have any cousins?” and then left as soon as they could. We also study just a small slice of the Earth’s population, and there’s no guarantee that what we discover about undergrads doing studies for extra credit, or Americans taking online surveys for pennies, or Chicago commuters striking up conversations for fruit, will generalize to the rest of humanity.

[Read: Stop being so self-conscious]

Second, social situations vary too much. People did have a surprisingly nice time talking on a train in Chicago, but the same might not be true at a grocery store in Tallahassee, or in a New York elevator. The outcome might depend on whom you’re talking with, or what you’re talking about, or whether you’ll end up getting a banana.

Studying all these different contexts would be like emptying the ocean one teaspoon at a time. At best, a few folks will run some additional studies fleshing out the phenomenon, and then everyone will move on. We will never have a truly comprehensive account of when, and for whom, talking to strangers is a good idea.

So what’s the point of all this research? Can you ever apply it to your daily life, or is it better to ignore it, like a sweaty interloper in an escape room?

My advice is to think of psychology research less as a set of instructions and more as a means of refining your intuitions. If you expect talking to strangers to be a terrible ordeal, then you should wonder why study participants find it surprisingly enjoyable. It’s possible those studies are wrong. But if they’re not, what gives? Maybe you’re just part of a minority of misanthropes. Maybe the strangers you meet aren’t like the strangers on that commuter line near Chicago. Maybe you treat every surprisingly delightful stranger as an exception and assume the next stranger will be bad.

[Read: How to talk to strangers]

Each new finding in psychology presents an opportunity to pick out the most useful bits, learn from them, and ignore the rest. We’re already used to doing this in other contexts. When we hear a narrative, we understand that some details matter (“Brutus betrayed Caesar”), and some don’t (“Brutus wore a toga”). We know that a story shows us what can happen (“Sometimes friends turn on you”), not what always happens (“Every friend will turn on you”). And we intuit that a story’s message should be taken seriously (“Make sure you maintain your friends’ loyalty”) and not literally (“Make sure to wear a stab-proof vest”). Nobody has to tell us how to reason in this way.

Applying our story sense to psychology works because psychology is stories. Each study reports what a certain group of people did in a certain time and place––that is, it sets a scene, fills it with characters, and puts them in motion. The stories can be simple (“People who said they felt depressed also said they had trouble sleeping”), or they can be complicated (“We offered people a banana to go talk to strangers on a train, and they reported having a better time than they expected”). We use statistics to show that our stories are credible, but a little bit of math doesn’t change what’s underneath.

I ended up ruining two perfectly good birthday parties because I didn’t use my story sense. The science at the conference in Atlanta suggested that meeting strangers can be unexpectedly wonderful, but I didn’t consider the context: Obviously, striking up a conversation with a fellow commuter is nothing like locking yourself in a room full of people who are trying to celebrate their friend’s birthday and enjoy a series of speakeasy-themed puzzles. I acted like the studies showed that conversations with strangers always go better than expected, rather than showing that they sometimes do. And I took the research literally (go meet new people right now!) rather than seriously (be more open to meeting new people, but, you know, don’t be stupid about it!).

A story sense can sometimes be misleading, though, as psychologists have shown in many different ways. For instance, people tend to assume that easily imagined events (such as dying in a terrorist attack) are more common than events that are hard to picture (such as dying from falling out of bed). Learning how to apply the findings of psychology research is not like learning long division or computer programming; there isn’t a handbook, and nobody can tell you when you’re doing it wrong. You pick it up slowly, painfully, through trial and error, when you see the crestfallen faces of the people whose birthdays you’ve ruined. No amount of expertise can speed up that process, which is why psychologists can study happiness and marriage all we want and yet some of us still end up depressed and divorced.

But here’s one finding you should take literally: Don’t sign up to do escape rooms with strangers. And if you’re one of the unfortunate people in those escape rooms I crashed: I’m so sorry, and please give my best to Cersei.

If you spend any time on the internet, you’re likely now familiar with the gray-and-teal screenshots of AI-generated text. At first they were meant to illustrate ChatGPT’s surprising competence at generating human-sounding prose, and then to demonstrate the occasionally unsettling answers that emerged once the general public could bombard it with prompts. OpenAI, the organization that is developing the tool, describes one of its biggest problems this way: “ChatGPT sometimes writes plausible-sounding but incorrect or nonsensical answers.” In layman’s terms, the chatbot makes stuff up. As similar services, such as Google’s Bard, have rushed their tools into public testing, their screenshots have demonstrated the same capacity for fabricating people, historical events, research citations, and more, and for rendering those falsehoods in the same confident, tidy prose.

This apparently systemic penchant for inaccuracy is especially worrisome, given tech companies’ intent to integrate these tools into search engines as soon as possible. But a bigger problem might lie in a different aspect of AI’s outputs—more specifically, in the polite, businesslike, serenely insipid way that the chatbots formulate their responses. This is the prose style of office work and email jobs, of by-the-book corporate publicists and LinkedIn influencers with private-school MBAs. The style sounds the same—pleasant, measured, authoritative—no matter whether the source (be it human or computer) is trying to be helpful or lying through their teeth or not saying anything coherent at all.

[Read: AI search is a disaster]

In the United States, this is the writing style of institutional authority, and AI chatbots are so far exquisitely capable of replicating its voice, while delivering information that is patently unreliable. On a practical level, this will pose challenges for people who must navigate a world with this kind of technology suddenly thrust into it. Our mental shortcuts used for evaluating communicative credibility on the fly have always been less than perfect, and the very nature of the internet already makes such judgment calls more difficult and necessary. AI could make them nearly impossible.

ChatGPT and its ilk are built using what are known as large language models, or LLMs. That means they hoover up very large quantities of written language online and then, very crudely speaking, analyze that data set to determine which words would likely be assembled in which order to create a successful response. They generate text that’s been optimized for plausibility, not for truthfulness. Being right isn’t the goal, at least not now; sounding right is. For any particular query, there are many more answers that sound right than answers that are true. LLMs aren’t intentionally lying—they are not alive, and cannot produce results meaningfully similar to human thought. And they haven’t been created to mislead their users. The chatbots do, after all, frequently generate answers that are both plausible and correct, even though any veracity is incidental. They are, in other words, masters of bullshit—persuasive speech whose essence “is just this lack of connection to a concern with truth—this indifference to how things really are,” the philosopher Harry Frankfurt wrote in his book-length essay on this sort of rhetoric.

[Read: Elon Musk, baloney king]

What LLMs are currently capable of producing is industrially scaled, industrial-grade bullshit. That’s troublesome for many reasons, not least of which is that humans have enough trouble discerning the age-old artisanal variety. Every human is required to make a zillion tiny decisions every day about whether some notion they’re presented with should be believed, and rarely do they have the opportunity or desire to stop, gather all the relevant information, and reason those decisions from first principles. To do so would pretty much halt human interaction as we know it, and even trying would make you pretty annoying.

So people instead rely on cognitive heuristics, which are little shortcuts that, in this case, help tip us toward belief or disbelief in situations where the full facts are unknown or unknowable. When you take medical advice from your doctor, you’ve employed an authority heuristic, which assigns trust in sources you believe have specialized knowledge and expertise. When you decide that something is probably true because it’s become the consensus among your family and friends, that’s the bandwagon heuristic at work. Even the best heuristics aren’t perfect: Your doctor might disbelieve your reported symptoms and misdiagnose you, or your social circle might be riddled with people who think the Earth is flat. But according to Miriam Metzger, a professor at UC Santa Barbara who studies how people evaluate credibility online, many of these shortcuts are, on balance, largely sound and extremely useful. Most people in most situations, for example, would be well served to listen to their doctor instead of taking medical advice from their weird cousin.

The growth of the internet has posed all kinds of issues for the accurate use of credibility heuristics, Metzger told me. There are too many potential sources of information vying directly for your attention, and too few ways to evaluate those sources or their motives quickly. Now your weird cousin is posting things on Facebook—and so are all of his weird friends, and their friends too. “The digital environment gives us a vastness of information in which it’s just harder for consumers to know who and what to trust,” Metzger said. “It’s put more of the burden on individuals to make their own credibility assessment practically every time they are confronted with new information.”

In the United States, this informational fragmentation is usually seen through the lens of politics, but it has also seeped into more mundane parts of life. On the internet, everyone can theoretically access expertise on everything. This freedom has some huge upsides, especially for people trying to solve small, manageable problems: There are enough instructional YouTube videos and Reddit threads to make you into your own travel agent, mechanic, plumber, and physical therapist. In many other scenarios, though, making judgment calls on the internet’s conglomeration of questionably sourced knowledge and maybe-faux expertise can have real consequences. We often don’t have anywhere near the information we’d need to evaluate a source’s credibility, and when that happens, we generally start rummaging through our bag of heuristics until we find one that works with whatever context we do have. What we end up with might just be the fluency heuristic—which is to say, the sense that certain patterns of communication are inherently credible.

In mainstream American culture, good grammar, accurate spelling, and a large and varied vocabulary free of expletives, slurs, or slang are all prerequisites for credibility, and a lack of them can be used to discredit challengers to existing authority and malign people with less education or different cultural backgrounds. This heuristic also can be easily used against the people who employ it: The more the phishing email looks and sounds like real communication from your bank, the more accounts scammers get to drain.

This is where the tidy, professional corporate-speak of well-trained LLMs has serious potential to cause informational chaos, Metzger said. Among other sources, the best AIs are trained on editorial content from major media organizations, archives of academic research, and troves of government and legal documents, according to a recent report by The Washington Post. These are just the type of source that would employ a precise and highly educated communication style. ChatGPT and other chatbots like it are text-generation machines that make up facts and sever information from its source. They are also authority-simulation machines that discourage readers from ever doubting them in the first place.

Whenever I want to avoid work, Google Maps is my go-to. The point is not to hunt for a bite to eat or plan a trip—it’s pure entertainment. I glide over a digital rendering of the Earth, spin it like a globe, and zoom in. Cities and rivers and streets and businesses begin to come into focus, colored with millions of user-uploaded images and reviews. It’s a bit like Where’s Waldo?, but at the world’s scale. Among the treasures to discover are blurry photographs of late nights in dive bars, ratings of obscure colonial-American-life museums, and selfies on a mountain I’ll never visit. There are the juiciest one-star restaurant reviews, such as this one from a café in my neighborhood: “totally unwarranted douche energy from ownership, expired motor oil passed off as coffee, and price-gouged food that is prepared and sourced as terribly as their coffee.” Brutal. In Ecuador, I found half a dozen businesses that appear to be Simpsons-themed.

Google Maps’ main purpose is to enable people to get directions and look up businesses. But along the way, it has become a social space too. Sort of. To fill out the world map it created, Google invited people to add snippets to all the digital places. You upload your photos; you leave your reviews; you look at the artifacts others have left behind. The pictures of a restaurant on Google Maps are often a mismatched succession of interior-design shots, flash photos of messy plates, and outdated menus. There’s plenty of detritus too: irrelevant photos, businesses that don’t exist, three-star reviews without an explanation.

The result is random and messy in a way that is different from the rest of the social web. Instagram and TikTok, after all, are dominated by hyper-curated influencer content, served through feeds geared precisely to specific tastes. Sometimes, pulling up TikTok’s “For You” page and getting sucked in by the app’s selections feels amazing. But especially as algorithmic content has taken over the web, many of the surprises don’t feel fresh. They are our kind of surprises. Google Maps offers something many other platforms no longer can: a hodgepodge of truly unfamiliar stuff that hasn’t been packaged for your taste or mine.

Not that Google necessarily deserves any praise for the delightful weirdness of Google Maps. The platform has become a refuge from the internet that Google itself had a heavy hand in sculpting. By turning its search engine into the internet’s front page—and deciding what appears on it—the company may be the culprit most responsible for ending unexpected encounters. Algorithms are also part of Google Maps: It does plenty of window dressing in its depictions of places, using its Your Match score to present establishments it anticipates you’ll enjoy. “It’s trying to match you with things that it thinks you’re going to like,” Kath Bassett, a sociologist at the University of Bath, in England, who studies map-based media platforms, told me. But “you can get past that.”

Because zooming out and scrolling around are so easy, you can bump into little treasures at every turn that would never land on an Instagram feed. Exploring Google Maps for kicks is not especially common, but I’m not the only one who does it: A community of people keen on these digital impressions has existed for some time. And GeoGuessr, a game in which players are dropped into a random location on Google Maps’ street view and try to guess where they are, has garnered a cult following.

Hopscotching around Google Maps unleashes surprise after surprise. According to one reviewer, the second-largest Union Jack in the British empire is at a museum called Cupids Legacy Centre, in Newfoundland, earning the facility four stars from the user. For only $6, one reviewer claims, you can “take in a tour of the local archeological dig site.” Seems like a steal. Several clicks away is the Tashkent Television Tower, in Uzbekistan’s capital, where scores of users have posted photos from the observation deck, some 300 feet up. The tower is well rated, but the café inside is not. “The service was slow. Our seats were wet,” wrote one visitor.

Thousands of miles west, in the Albanian town of Elbasan, a man in purple poses coolly—hands pocketed, sunglasses activated—by the coat rack at “Supermarket Koli.” A moment later, I am out in the Delaware Bay, many miles away from land, where the Fourteen Foot Bank Lighthouse juts out of the ocean. A few people have been kind enough to upload images of the 136-year-old haunted metal hulk; how else would I possibly encounter this thing? The strange moments you dig up are typically encased in blandness—drab photos, anodyne reviews, useless information that bores—but that contrast is what gives the weird little bits their value, like a diamond enshrined in stone.

Still, Google Maps hardly presents an entirely accurate depiction of our world. Fewer businesses seem to be indexed in rural areas and the global South, and Street View mode becomes more sparse. Reality is sometimes ugly, and to some degree, the platform moderates its content. According to its website, content reflecting more abrasive realities—bigotry, violence, anything sexually explicit—is pulled, often abetted by machine learning. In some cases, Google Maps itself affects reality: The platform has assumed so much power that its decisions about what to name neighborhoods can have major ramifications.

But as a repository of moments, it can offer depth that other sites, and even perhaps reality, cannot. Happenstance moments at my laundromat posted online aren’t the same as the ones I experience when I’m out of socks, but they give the place a greater dimensionality. It becomes more vibrant in my mind—even if by way of confusing, warped photos of rickety dryers. Other sites, such as Reddit and Wikipedia, are also full of odd curiosities, with devoted fans, but none lets such a random potpourri collect with easy visibility for users. Whereas Wikipedia has a base of meticulous contributors fleshing out structures of information, Google Maps is flavored by masses of people from everywhere throwing up ad hoc content—its ubiquity unlocks possibilities. More than 1 billion people use Google Maps every month.

Compare that with the rest of the social web: Any possible travel destination, for example, is now paired with an endless supply of TikTok videos and Instagram foodie listicles. Look and think no further; we have your Mexico City trip all planned out. Such a polished depiction of a place is sometimes genuinely helpful, but predicting what will resonate with us can be difficult. And what does might be too small or unconventional for selection by an algorithm or an influencer. Because it’s so huge, Google Maps has you looking under a lot of rocks for something good—more so than a “For You” feed delivering amusing content into your hands. But when you uncover something special on Google Maps, it’s like finding $100 on the sidewalk. You so easily could have missed it, but you found it. That kind of satisfaction is tough to replicate on social media today. Recently, I spent 20 minutes looking at this picture of a troop of Siberian teenagers in red berets watching an old man ambiguously stand next to an orange hazmat suit. What is happening here?

Most people won’t end up treating Google Maps as a form of unfiltered entertainment. But even using the platform in the most functional way can influence how you perceive the world. Looking up a pharmacy or the nearest rest stop, you can’t help but bump into off-kilter stuff other users have shared. Google Maps remains an open door to “strange places where all kinds of things are still happening,” Christian Sandvig, a professor at the University of Michigan’s School of Information who studies algorithms, told me. “There was a period where everyone was excited about a kind of shaggier internet, where everyone could put stuff up.” For better or for worse, “it was this free-for-all. And it definitely feels like it’s going away,” he said. The site carries on the legacy of an earlier internet tasked with connecting the world to itself rather than pushing all of us into silos. It is among the internet’s last bastions of unfiltered weirdness.

On a recent Google Maps scrolling expedition, I ended up in Westernpark Nemesvita, a cheesy, artificial little cowboy town near the northern shores of Lake Balaton, Hungary’s largest body of water. There’s archery, a small trampoline, and an inflatable water slide labeled Grand Canyon. A peacock appears to walk the grounds. One glowing review reads, “If you are in Hungary you have to stop by there.” Maybe one day.

As the world marked the 52nd edition of Earth Day, thousands took to the streets in Montreal to tell the Canadian government to do more concerning climate change.

Climate change campaigners mark Earth Day with protest focusing on nature and biodiversity.

Some protesters lay down for a "die-in" to show "humans will not survive if nothing is done about climate change".

Peruvian shamans perform rituals on Earth Day