Earlier today, three witnesses came before Congress to testify about their experiences with unidentified flying objects. A former Navy pilot spoke of the mysterious objects that he has seen with his own eyes and through radar, and how frequently pilots encounter them in the air. A retired Navy commander described the time he pulled his jet up to an object shaped like a Tic Tac hovering over the ocean, then watched it suddenly speed up and vanish.

The most anticipated remarks, however, came from a former military-intelligence officer named David Grusch, who went public with his account just last month. Grusch told the House oversight subcommittee on national security that the American government has spent decades secretly recovering mysterious vehicles that have crashed on the ground, and has determined the material to be of “non-human” origin. The government also attempted to reverse engineer some of the technology. And it’s doing all of this clandestinely, without proper supervision by Congress.

In the hearing, Grusch expanded on his previous claims in response to lawmakers’ questions. If elected officials had never heard about this effort before, how did it get any funding? The military pilfered money that had been allocated for its other programs. A defense official recently testified before Congress that the U.S. military hasn’t found any evidence of extraterrestrial activity on Earth; is that statement correct? It’s not accurate. Has any of the activity been aggressive or hostile? My colleagues have gotten physically injured. By UFOs, or people within the government? Both.

After not holding a hearing on UFOs for more than half a century, Congress has recently held two in as many years. In that sense, we can count today’s events as historic. But as in the other hearings, this one had no big reveal, no grand answer to humankind’s most existential questions about our place in the universe. The hype surrounding the hearing—and there has been considerable hype—says more about the people who tuned in than about Grusch’s claims. Just as it did in the late 1940s, when stories of flying saucers over Washington state and crash-landings in New Mexico captivated the nation, UFO fever today indicates that Americans feel that their government knows more than it’s letting on.

That sentiment is not new, nor is Americans’ belief in conspiracy theories. Though research suggests that conspiracy thinking is not getting worse in the modern-day United States, we are in a moment of acute public curiosity about—and acceptance of—conspiracism. Compared with QAnon, vaccine microchips, and stolen elections, a big UFO cover-up might seem almost reasonable. Even if that cover-up involves, as Grusch previously claimed in an interview, the military discovering the “dead pilots” of alien craft. (In Congress today, Grusch declined to give specifics about this and many other claims, saying that there was only so much he could disclose to the public and that he could elaborate in a closed setting.)  

The past several years have coincided with an unprecedented mainstreaming of UFO culture. In 2017, when an interstellar object showed up in our solar system, most scientists agreed that it was an asteroid or a comet, but some said it could have been an alien spaceship. (The Harvard professor leading the latter camp, Avi Loeb, recently led an expedition to recover material from the seafloor that he believes could be from alien spacecraft.) Later that year, The New York Times and other news outlets revealed that the Pentagon had a covert program dedicated to cataloging UFOs. Then NASA decided to weigh in on the topic after years of steering clear, and convened a team to consider UFOs in a “scientific perspective.” And who can forget the spy balloons that the military shot out of the sky this year?

These events have unfolded against a shift in public knowledge about the universe beyond Earth, which might help explain why people are interested. In the 1940s, the only planets we knew of were the ones around our sun, and scientists had only recently determined that there were galaxies other than our own. Today, astronomers have discovered more than 5,000 exoplanets, and telescopes can see nearly all the way back to the Big Bang. Faced with so many wonders, the question of whether we’re sharing them with anyone else becomes more urgent, and might even seem more answerable. “I think people are just ready, or at least excited about the possibilities of alien contact, maybe more than ever,” Jacob Haqq Misra, an astrobiologist at the Blue Marble Space Institute of Science, told me.

Congress has contributed to this mainstreaming too. Under the instruction of lawmakers, the Pentagon last year established a special office dedicated to investigating reports of unexplainable phenomena in the sky, at sea, and on land. The effort has been unusually bipartisan, with both far-right Republicans and progressive Democrats calling on the military to be more transparent. This month, Senator Chuck Schumer introduced legislation that would create a commission with the authority to declassify government documents about UFOs. “The American public has a right to learn about technologies of unknown origins, non-human intelligence, and unexplainable phenomena,” Schumer said in a statement.

Yes, we do. But some undisclosed documents about UFOs is not synonymous with incontrovertible evidence that aliens have visited Earth. UFOs are just that—objects that are flying, and that we cannot yet identify. If the military is misusing taxpayer money to investigate mysterious debris it doesn’t recognize, that’s bad, whether it’s the remnants of drones from another nation or a non-human craft. “If that’s the case, and auditors have not been allowed into these programs and there’s illegal layers of secrecy,” Haqq Misra said, “then that’s really important to disclose, independent of any connection to anything else”—anything otherworldly. But even as lawmakers assert that UFOs are primarily a national-security concern, by invoking aliens in their discussions, they lend credence to the idea that a connection between the two exists.

Grusch was careful to tell lawmakers that he was only “speaking to the facts as I have been told them”—that is, he has not seen any evidence of alien wreckage or its inhabitants himself. And in general, though his claims are steeped in the language of authority, he simply has not been able to offer any concrete proof. The news website that first published Grusch’s claims reported that the Pentagon had cleared him to speak publicly, but that means only that his remarks don’t contain classified information, not that they’re true. Testifying under oath before Congress is not a measure of truth, either. Outside the hearing, some lawmakers seemed like they didn’t know what to make of the claims.

The prospect of extraterrestrial interlopers may be a national-security question, but it’s also a scientific one. Science requires data, and secondhand accounts just aren’t data. “When NASA brings back rocks on the moon, those rocks are shared with qualified people,” David Spergel, an astrophysicist at Princeton who chaired NASA’s committee on UFOs, told me. “Imagine we had some samples of some craft [and] we really want to understand what it was. You would make materials from those small samples available for labs anywhere in the world.” In other words, meaningful testimony would show evidence of alien ships and pilots, not just tell the public about them. “That would be pretty awesome,” he said, but it’s not what we’ve got. Today, we heard some extraordinary claims, and, to quote Carl Sagan, they require extraordinary evidence.

Last week, at a Fox News town hall (where else?), former President Donald Trump called China’s despot, Xi Jinping, a “brilliant” guy who “runs 1.4 billion people with an iron fist.” Lest anyone doubt his admiration, Trump added that Xi is “smart, brilliant, everything perfect. There’s nobody in Hollywood like this guy.”

Trump is not alone. Many in the United States and around the globe see the allure of a dictator who gets things done and makes the trains run on time, no matter the rules or laws that stand in the way. According to repeated polling, roughly one in four Americans agrees with the statement that a “strong leader who doesn’t have to bother with Congress and elections” is desirable. A much higher proportion of citizens agrees with that sentiment elsewhere, including in some of the most populous democracies: 55 percent of Indians, 52 percent of Indonesians, 38 percent of Nigerians, and 31 percent of Japanese.

This grass-is-greener view of authoritarian rule tends to emerge most often where governments are failing to meet popular expectations. When democracy delivers, dictatorship doesn’t seem like a rosy alternative. Only 6 percent of Germans and 9 percent of Swedes are seduced by strongmen.

[Brian Klaas: Democracy has a customer-service problem]

Admiration for autocracy is built on a pernicious lie that I call the “myth of benevolent dictatorship.” The myth is built on three flimsy pillars: first, that dictators produce stronger economic growth than their democratic counterparts; second, that dictators, unswayed by volatile public opinion, are strategic long-term thinkers; and third, that dictators bring stability, whereas divided democracies produce chaos.  

Two decades ago, the United States and its Western allies became embroiled in Iraq and later blundered into the financial crisis, leading think tanks to begin praising the “Beijing Consensus,” or the “China Model,” as an alternative to liberal democracy. Critiques of democracy surged in popularity in the era of Trump and Brexit. In the United States, intellectual publications ran articles arguing that the problem was too much democracy. In 2018, The Times of London published a column titled “Our Timid Leaders Can Learn From Strongmen.” China’s state media, capitalizing on the West’s democratic woes, argued that democracy is a “scary” system that produces self-inflicted wounds.

But events and new research in the past several years have taken a wrecking ball to the long-standing myth of benevolent dictatorship. All three pillars of the lie are crumbling. Every fresh data point proves Winston Churchill right: “Democracy is the worst form of Government, except for all those other forms that have been tried from time to time.”

Let’s start with the myth that dictatorships produce stronger growth. This falsehood arose from a few well-known, cherry-picked examples, in which despots oversaw astonishing transformations of their national economy. Starting in the late 1950s, Lee Kuan Yew helped transform Singapore from a poor, opium-filled backwater into a wealthy economic powerhouse. And in China, per capita GDP rose from nearly $318 in 1990 to more than $12,500 today. Those successes are eye-popping.

But a systematic evaluation of the overall data reveals another reality. Even with these outliers of strong growth, most rigorous studies have found limited or no evidence that authoritarian regimes produce better economic growth than democratic ones. Some researchers, such as the political economists Darren Acemoglu and James Robinson, have found compelling evidence that the inclusive political institutions of democracy are one of the strongest factors in producing stable, long-term growth.

When authoritarian regimes do succeed economically, they often do so at a cost, because even booming dictatorships are prone to catastrophic busts. As the political scientist Jacob Nyrup has written: “China has within a 50-year time frame both experienced a famine, where 20-45 million people died, and an economic boom, where hundreds of millions of people were lifted out of poverty.” The rosiest interpretation of the authoritarian economic data, then, is that autocrats may sometimes preside over marginally higher growth, but with a much greater risk of economic collapse. That’s not a wise trade-off.

However, the myth of strongmen as economic gurus has an even bigger problem. Dictators turn out to have manipulated their economic data for decades. For a long time, they’ve fooled us. But now we have proof: The reason their numbers sometimes seem too good to be true is that they are.

Every government has motivation to fudge its economic data. But democracies have institutions that provide oversight and block politicians from that impulse, ensuring accurate figures. No such checks exist in dictatorships.

That difference led Luis Martinez, an economist at the University of Chicago, to test whether despots were overstating their growth rate. He did so with an ingenious method. Previous studies have verified the presence of a strong, accurate correlation between the amount of nighttime light captured by satellites and overall economic activity. When economies grow, they emit more nighttime light (which is why you can clearly pick out cities on a nighttime satellite image, and why the density of light is so much lower in Africa than, say, in Europe or on the American East Coast). High-resolution images allow researchers to track changes in nighttime illumination over time, and the detailed, granular data these images produce are nearly impossible to manipulate. Martinez discovered an astonishing disparity suggesting that dictators have been overstating their GDP growth by about 35 percent.

And the more the numbers are checked, the more manipulation is exposed. In Rwanda, where The New York Times has named President Paul Kagame “the global elite’s favorite strongman” because of his apparently brilliant record of economic growth, the government claimed that it had decreased poverty by 6 percent from 2010 to 2014. Researchers found that the inverse was true: Poverty had actually surged by 5 to 7 percent. Fittingly, the notion that Benito Mussolini made the trains run on time was a lie; he built ornate stations and invested in train lines used by elites, but the commuting masses got left behind.

[Read: The undoing of China’s economic miracle]

Even China, the apparent authoritarian economic miracle, is showing signs of slowing down, its growth model no longer so well matched to the global economy. Such cracks in growth are an innate feature of autocracy. Because dictatorships criminalize dissent, normal mechanisms of economic feedback are broken, and the system doesn’t self-correct when blundering into economic mistakes. Beijing’s quixotic quest to maintain perpetual “zero COVID” was a case in point. Autocrats are adept at building ports and roads and mines. But thriving modern economies are sustained less by open mines than by open minds, of which dictatorships, by design, have a limited supply.

Advocates for the myth of benevolent dictatorship conveniently ignore a crucial fact, which is that much of the growth in autocracies comes either from manufacturing products that were invented in the more open societies of the democratic West, or from exporting goods to rich democracies. (The top destinations for Chinese exports are the United States, Japan, and South Korea.) In that way, even the outliers of autocratic growth depend for their success on the innovation and consumer wealth of democracies. Would China have lifted millions out of poverty through export-led growth quite so fast if democratic America hadn’t become an economic powerhouse first?

The myth’s second pillar turns out to be no less rickety than the first. It holds that dictators are more strategic long-term thinkers than democrats because they’re not beholden to fickle public opinion. But this lie is believable only if you don’t understand how most dictatorships actually work.

Over more than a decade, I’ve studied and interviewed despots and the henchmen who surround them. One conclusion I’ve drawn is that making decisions based on bad information is an intrinsic feature of the systems dictators run. The longer despots cling to power, the more likely they are to fall into what I call “the dictator trap,” in which they crush dissent, purge anyone who challenges them, and construct their own reality through propaganda, all to maintain control. Speaking truth to power in such a system can literally be deadly. As a result, dictators are told only what they want to hear, not what is true, and they begin to believe their own lies. Vladimir Putin’s catastrophic war in Ukraine is a tragic illustration of the dictator trap: Putin got high on his own supply, and innocent Ukrainians are the victims of his power trip.

Despots often use their power not for long-term planning, but for short-term self-glorification, as no end of examples can attest. Turkmenistan’s former dictator Saparmurat Niyazov blew millions to build, in his own honor, a golden statue that would rotate to always face the sun. In another stroke of genius, he closed all rural hospitals so that the sick could have the privilege of being treated in his pristine marble capital of Ashgabat. Most of the population lived outside the city, and countless thousands likely died because they couldn’t reach a hospital in time. His successor erected an enormous golden statue of his favorite breed of dog. Thankfully, democracies have checks and balances to suppress such narcissistic whims.

The most persistent pillar of the myth, however, is the one that holds that dictators produce stability. Some dictators have hung on to power for decades. Before his death, Muammar Qaddafi ruled Libya for 42 years. Paul Biya of Cameroon, an 89-year-old despot who had no idea where he was during a recent event, took office during the Vietnam War. Putin has been in power for more than two decades; Xi has ruled for only one so far, but he appears prepared to retain his position indefinitely.

To stay in power, authoritarian leaders face constant trade-offs. If they strengthen military or paramilitary leaders, they face the risk of a coup d’état. But if they weaken their men under arms, then they can’t protect themselves from external invasion. To keep their elites happy, despots need to make them rich through corruption—usually at the expense of the population. But a ruling class awash in ill-gotten gains could inspire a revolution, or a wild card: assassination. Autocrats appear stable, but they’re not. They’re constantly vulnerable, forced to make every decision based on what will stave off threats to survive in power.

The stability that does exist in autocracies is, ironically, derived partially from the trappings of democracy. Recent research has made clear that dictators have developed mechanisms to “mimic democracy to prolong autocracy.” Most authoritarian leaders now hold elections, but rig them. Some use parliaments or courts to enact unpopular decisions while avoiding blame.

[From the December 2021 issue: The bad guys are winning]

Eventually, though, dictatorships tend to fall apart. And when they collapse, they really collapse. Elections in democracies change governments, not regimes. Personalist dictatorships, by contrast, often implode. When Qaddafi was killed, Libya disintegrated. He had deliberately designed the political system to function only with him at its center. The same could be true of Putin’s Russia. When he is toppled or dies, the country won’t have a smooth, peaceful transition.

The often-disastrous demise of autocrats creates a negative feedback loop. Nearly seven in 10 leaders of personalist dictatorships end up jailed, exiled, or killed once they lose power. While in power, many despots are aware of this grim fact, and so they use violence to stay in power, often growing more extreme as they lurch toward their downfall. The effect can hardly be called “stability,” even if the same person occupies the palace for decades.

For anyone who still clings to the illusion that dictatorships are likely to be prosperous, strategically wise, or internally stable, I propose a simple test. Imagine that someone wrote down the names of all the countries in the world on little slips of paper and then separated them into two hats: one for democracy, one for dictatorships. You would select one of the two hats, draw a slip of paper from it, look at the name, and then spend the rest of your life living in that country. Who knows, maybe you’d get lucky and end up in an authoritarian regime that seems stable and is producing steady growth. But I know which hat I would choose. And even if you fantasize about finding the unicorn that is a benevolent strongman, I suspect you do too.

On weekdays, between homeschooling her two children, Michelle Curtis logs on to her computer to squeeze in a few hours of work. Her screen flashes with Google Search results, the writings of a Google chatbot, and the outputs of other algorithms, and she has a few minutes to respond to each—judging the usefulness of the blue links she’s been provided, checking the accuracy of an AI’s description of a praying mantis, or deciding which of two chatbot-written birthday poems is better. She never knows what she will have to assess in advance, and for the AI-related tasks, which have formed the bulk of her work since February, she says she has little guidance and not enough time to do a thorough job.

Curtis is an AI rater. She works for the data company Appen, which is subcontracted by Google to evaluate the outputs of the tech giant’s AI products and search algorithm. Countless people do similar work around the world for Google; the ChatGPT-maker, OpenAI; and other tech firms. Their human feedback plays a crucial role in developing chatbots, search engines, social-media feeds, and targeted-advertising systems—the most important parts of the digital economy.

Curtis told me that the job is grueling, underpaid, and poorly defined. Whereas Google has a 176-page guide for search evaluations, the instructions for AI tasks are relatively sparse, she said. For every task she performs that involves rating AI outputs, she is given a few sentences or paragraphs of vague, even convoluted instructions and as little as just a few minutes to fully absorb them before the time allotted to complete the task is up. Unlike a page of Google results, chatbots promise authoritative answers—offering the final, rather than first, step of inquiry, which Curtis said makes her feel a heightened moral responsibility to assess AI responses as accurately as possible. She dreads these timed tasks for the very same reason: “It’s just not humanly possible to do in the amount of time that we’re given.” On Sundays, she works a full eight hours. “Those long days can really wear on you,” she said.

Armughan Ahmad, Appen’s CEO, told me through a spokesperson that the company “complies with minimum wages” and is investing in improved training and benefits for its workers; a Google spokesperson said Appen is solely responsible for raters’ working conditions and job training. For Google to mention these people at all is notable. Despite their importance to the generative-AI boom and tech economy more generally, these workers are almost never referenced in tech companies’ prophecies about the ascendance of intelligent machines. AI moguls describe their products as forces akin to electricity or nuclear fission, like facts of nature waiting to be discovered, and speak of “maximally curious” machines that learn and grow on their own, like children. The human side of sculpting algorithms tends to be relegated to opaque descriptions of “human annotations” and “quality tests,” evacuated of the time and energy powering those annotations.

[Read: Google’s new search tool could eat the internet alive]

The tech industry has a history of veiling the difficult, exploitative, and sometimes dangerous work needed to clean up its platforms and programs. But as AI rapidly infiltrates our daily lives, tensions between tech companies framing their software as self-propelling and the AI raters and other people actually pushing those products along have started to surface. In 2021, Appen raters began organizing with the Alphabet Workers Union-Communications Workers of America to push for greater recognition and compensation; Curtis joined its ranks last year. At the center of the fight is a big question: In the coming era of AI, can the people doing the tech industry’s grunt work ever be seen and treated not as tireless machines but simply as what they are—human?  

The technical name for the use of such ratings to improve AI models is reinforcement learning with human feedback, or RLHF. OpenAI, Google, Anthropic, and other companies all use the technique. After a chatbot has processed massive amounts of text, human feedback helps fine-tune it. ChatGPT is impressive because using it feels like chatting with a human, but that pastiche does not naturally arise through ingesting data from something like the entire internet, an amalgam of recipes and patents and blogs and novels. Although AI programs are set up to be effective at pattern detection, they “don’t have any sense of contextual understanding, no ability to parse whether AI-generated text looks more or less like what a human would have written,” Sarah Myers West, the managing director of the AI Now Institute, an independent research organization, told me. Only an actual person can make that call.

The program might write multiple recipes for chocolate cake, which a rater ranks and edits. Those evaluations and examples will inform the chatbot’s statistical model of language and next-word predictions, which should make the program better at writing recipes in the style of a human, for chocolate cake and beyond. A person might check a chatbot’s response for factual accuracy, rate how well it fits the prompt, or flag toxic outputs; subject experts can be particularly helpful, and they tend to be paid more.

Using human evaluations to improve algorithmic products is a fairly old practice at this point: Google and Facebook have been using them for almost a decade, if not more, to develop search engines, targeted ads, and other products, Sasha Luccioni, an AI researcher at the machine-learning company Hugging Face, told me. The extent to which human ratings have shaped today’s algorithms depends on who you ask, however. Major tech companies that design and profit from search engines, chatbots, and other algorithmic products tend to characterize the raters’ work as only one among many important aspects of building cutting-edge AI products. Courtenay Mencini, a Google spokesperson, told me that “ratings do not directly impact or solely train our algorithms. Rather, they’re one data point … taken in aggregate with extensive internal development and testing.” OpenAI has emphasized that training on huge amounts of text, rather than RLHF, accounts for most of GPT-4’s capabilities.

[From the September 2023 issue: Does Sam Altman know what he’s creating?]

AI experts I spoke with outside these companies took a different stance. Targeted human feedback has been “the single most impactful change that made [current] AI models as good as they are,” allowing the leap from GPT-2’s half-baked emails to GPT-4’s convincing essays, Luccioni said. She and others argue that tech companies intentionally downplay the importance of human feedback. Such obfuscation “sockets away some of the most unseemly elements of these technologies,” such as hateful content and misinformation that humans have to identify, Myers West told me—not to mention the conditions the people work under. Even setting aside those elements, describing the extent of human intervention would risk dispelling the magical and marketable illusion of intelligent machines—a “Wizard of Oz effect,” Luccioni said.

Despite tech companies’ stated positions, digging into their own press statements and research papers about AI reveals that they frequently do acknowledge the value of this human labor, if in broad terms. A Google blog post promoting a new chatbot last year, for instance, said that “to create safer dialogue agents, we need to be able to learn from human feedback.” Google has similarly described human evaluations as necessary to its search engine. The company touts RLHF as “particularly useful” for applying its AI services to industries such as health care and finance. Two lead researchers at OpenAI similarly described human evaluations as vital to training ChatGPT in an interview with MIT Technology Review. The company stated elsewhere that GPT-4 exhibited “large improvements” in accuracy after RLHF training and that human feedback was crucial to fine-tuning it. Meta’s most recent language model, released this week, relies on “over 1 million new human annotations,” according to the company.

To some extent, the significance of humans’ AI ratings is evident in the money pouring into them. One company that hires people to do RLHF and data annotation was valued at more than $7 billion in 2021, and its CEO recently predicted that AI companies will soon spend billions of dollars on RLHF, similar to their investment in computing power. The global market for labeling data used to train these models (such as tagging an image of a cat with the label “cat”), another part of the “ghost work” powering AI, could reach nearly $14 billion by 2030, according to an estimate from April 2022, months before the ChatGPT gold rush began.

All of that money, however, rarely seems to be reaching the actual people doing the ghostly labor. The contours of the work are starting to materialize, and the few public investigations into it are alarming: Workers in Africa are paid as little as $1.50 an hour to check outputs for disturbing content that has reportedly left some of them with PTSD. Some contractors in the U.S. can earn only a couple of dollars above the minimum wage for repetitive, exhausting, and rudderless work. The pattern is similar to that of social-media content moderators, who can be paid a tenth as much as software engineers to scan traumatic content for hours every day. “The poor working conditions directly impact data quality,” Krystal Kauffman, a fellow at the Distributed AI Research Institute and an organizer of raters and data labelers on Amazon Mechanical Turk, a crowdsourcing platform, told me.

Stress, low pay, minimal instructions, inconsistent tasks, and tight deadlines—the sheer volume of data needed to train AI models almost necessitates a rush job—are a recipe for human error, according to Appen raters affiliated with the Alphabet Workers Union-Communications Workers of America and multiple independent experts. Documents obtained by Bloomberg, for instance, show that AI raters at Google have as little as three minutes to complete some tasks, and that they evaluate high-stakes responses, such as how to safely dose medication. Even OpenAI has written, in the technical report accompanying GPT-4, that “undesired behaviors [in AI systems] can arise when instructions to labelers were underspecified” during RLHF.

Tech companies have at times responded to these issues by stating that ratings are not the only way they check accuracy, that humans doing those ratings are paid adequately based on their location and afforded proper training, and that viewing traumatic materials is not a typical experience. Mencini, the Google spokesperson, told me that Google’s wages and benefits standards for contractors do not apply to raters, because they “work part-time from home, can be assigned to multiple companies’ accounts at a time, and do not have access to Google’s systems or campuses.” In response to allegations of raters seeing offensive materials, she said that workers “select to opt into reviewing sensitive content, and can opt out freely at any time.” The companies also tend to shift blame to their vendors—Mencini, for instance, told me that “Google is simply not the employer of any Appen workers.”  

[Read: The coming humanist renaissance]

Appen’s raters told me that their working conditions do not align with various tech companies’ assurances—and that they hold Appen and Google responsible, because both profit from their work. Over the past year, Michelle Curtis and other raters have demanded more time to complete AI evaluations, benefits, better compensation, and the right to organize. The job’s flexibility does have advantages, they told me. Curtis has been able to navigate her children’s medical issues; another Appen rater I spoke with, Ed Stackhouse, said the adjustable hours afford him time to deal with a heart condition. But flexibility does not justify low pay and a lack of benefits, Shannon Wait, an organizer with the AWU-CWA, told me; there’s nothing flexible about precarity.

The group made headway at the start of the year, when Curtis and her fellow raters received their first-ever raise. She now makes $14.50 an hour, up from $12.75—still below the minimum of $15 an hour that Google has promised to its vendors, temporary staff, and contractors. The union continued raising concerns about working conditions; Stackhouse wrote a letter to Congress about these issues in May. Then, just over two weeks later, Curtis, Stackhouse, and several other raters received an email from Appen stating, “Your employment is being terminated due to business conditions.”

The AWU-CWA suspected that Appen and Google were punishing the raters for speaking out.  “The raters that were let go all had one thing in common, which was that they were vocal about working conditions or involved in organizing,” Stackhouse told me. Although Appen did suffer a drop in revenue during the broader tech downturn last year, the company also had, and has, open job postings. Four weeks before the termination, Appen had sent an email offering cash incentives to work more hours and meet “a significant spike in jobs available since the beginning of year,” when the generative-AI boom was in full swing; just six days before the layoffs, Appen sent another email lauding “record-high production levels” and re-upping the bonus-pay offer. On June 14, the union filed a complaint with the National Labor Relations Board alleging that Appen and Google had retaliated against raters “by terminating six employees who were engaged in protected [labor] activity.”

Less than two weeks after the complaint was filed, Appen reversed its decision to fire Curtis, Stackhouse, and the others; their positions were reinstated with back pay. Ahmad, Appen’s CEO, told me in an email that his company bases “employment decisions on business requirements” and is “happy that our business needs changed and we were able to hire back the laid off contributors.” He added, “Our policy is not to discriminate against employees due to any protected labor activities,” and that “we’ve been actively investing in workplace enhancements like smarter training, and improved benefits.”

Mencini, the Google spokesperson, told me that “only Appen, as the employer, determines their employees’ working conditions,” and that “Appen provides job training for their employees.” As with compensation and training, Mencini deflected responsibility for the treatment of organizing workers as well: “We, of course, respect the labor rights of Appen employees to join a union, but it’s a matter between them and their employer, Appen.”

That AI purveyors would obscure the human labor undergirding their products is predictable. Much of the data that train AI models is labeled by people making poverty wages, many of them located in the global South. Amazon deliveries are cheap in part because working conditions in the company’s warehouses subsidize them. Social media is usable and desirable because of armies of content moderators also largely in the global South. “Cloud” computing, a cornerstone of Amazon’s and Microsoft’s businesses, takes place in giant data centers.

AI raters might be understood as an extension of that cloud, treated not as laborers with human needs so much as productive units, carbon transistors on a series of fleshly microchips—objects, not people. Yet even microchips take up space; they require not just electricity but also ventilation to keep from overheating. The Appen raters’ termination and reinstatement is part of “a more generalized pattern within the tech industry of engaging in very swift retaliation against workers” when they organize for better pay or against ethical concerns about the products they work on, Myers West, of the AI Now Institute, told me.

Ironically, one crucial bit of human labor that AI programs have proved unable to automate is their own training. Human subjectivity and prejudice have long migrated their way into algorithms, and those flaws mean machines may not be able to perfect themselves. Various attempts to train AI models with other AI models have bred further bias and worsened performance, though a few have shown limited success. “I can’t imagine that we will be able to replicate [human intervention] with current AI approaches,” Hugging Face’s Luccioni told me in an email; Ahmad said that “using AI to train AI can have dire consequences as it pertains to the viability and credibility of this technology.” The tech industry has so far failed to purge the ghosts haunting its many other machines and services—the people organizing on warehouse floors, walking out of corporate headquarters, unionizing overseas, and leaking classified documents. Appen’s raters are proving that, even amid the generative-AI boom, humanity may not be so easily exorcized.

When Megan Nolan published her first novel, fellow authors warned her in “ominous tones” about the website Goodreads. The young Irish writer looked at the book’s listing there in the winter of 2020, the day the first proof copy arrived at her house. “Nobody but me and the publisher had seen it,” she wrote recently. “Despite this, it had received one review already: two stars, left by someone I had inconsequential personal discord with. It was completely impossible for him to have read the book.”

The terrible power of Goodreads is an open secret in the publishing industry. The review site, which Amazon bought in 2013, can shape the conversation around a book or an author, both positively and negatively. Today’s ostensible word-of-mouth hits are more usually created online, either via Goodreads or social networks such as Instagam and TikTok.

Publishers know how important these dynamics are, and so they send out advanced reading copies, or ARCs, not just to independent booksellers who might stock a title, but also to influencers who might make content about it. “There’s an assumption that if you receive an ARC that you will post about it,” Traci Thomas, host of the literary podcast The Stacks, told me—“whether that’s on your Goodreads, on your Instagram, on your TikTok, tell other people in your bookstore, or whatever. And so that’s how it ends up that there’s so many reviews of a book that’s not out yet.”

Many book bloggers are conscientious about including a disclaimer on their posts thanking the publisher for giving them an ARC “in exchange for an honest review.” But disclosing freebies is far from a contractual requirement or even a social norm. So you can’t easily discern which early reviewers have actually read the book, and which ones might be reacting to social-media chatter (or, as Nolan suspected in her case, prosecuting a personal grudge).

That matters because viral campaigns target unpublished books all the time. What tends to happen is that one influential voice on Instagram or TikTok deems a book to be “problematic,” and then dozens of that person’s followers head over to Goodreads to make the writer’s offense more widely known. Authors who reply to these attacks risk making the situation worse. Kathleen Hale—who was so infuriated by a mean reviewer that she tracked down the woman’s address—wrote later that the site had warned her against engagement: “At the bottom of the page, Goodreads had issued the following directive (if you are signed in as an author, it appears after every bad review of a book you’ve written): ‘We really, really (really!) don’t think you should comment on this review, even to thank the reviewer.’” Most authors I know read their Goodreads reviews, and then silently fume over them alone. Because I am a weirdo, I extract great enjoyment from mine—the more petty and baffling the complaints, the better. “I listened to the audiobook and by chapter 3 it started to annoy me the little pause she made before the word ‘male,’” reads one review of my book, Difficult Women.

When the complaints are more numerous and more serious, it’s known as “review-bombing” or “brigading.” A Goodreads blitzkrieg can derail an entire publication schedule, freak out commercial book clubs that planned to discuss the release, or even prompt nervous publishers to cut the marketing budget for controversial titles. Last month, the Eat, Pray, Love author Elizabeth Gilbert withdrew her upcoming novel The Snow Forest from publication because of the backlash she received after revealing it was set in Soviet Russia. The Goodreads page for The Snow Forest, which has since been taken down, accused her of romanticizing the Russian soul. “I’ll cut the job for you—they don’t have any,” wrote one reviewer. Another wrote: “Just like her characters in this nover [sic] are unaware of the events of WWII, Elizabeth Gilbert herself seems to be unaware of the genocidal war russia is conducting against Ukraine RIGHT NOW, because I’m sure if she knew, she’d realise how tone deaf this book is.”

[Read: Eat, pray, pander]

The book had been scheduled for release next February, but in a video announcing that it was “not the time for this book to be published,” Gilbert essentially endorsed the Goodreads criticisms: “I do not want to add any harm to a group of people who have already experienced and who are all continuing to experience grievous and extreme harm.”

Now, I don’t know whether The Snow Forest romanticized the Russian soul or would somehow have caused “harm” to Ukrainians. Like my colleague Franklin Foer, I find the allegations hard to believe. But the plain fact is that neither of us know, because—and this should be obvious, although recent events suggest it is not—you don’t know what’s in a book you haven’t read. You also don’t know what’s in a film you haven’t watched, an album you haven’t heard, or an article you haven’t clicked on. That used to matter. It no longer does, because we live in a world where you can harvest likes by circulating screenshots of headlines and out-of-context video clips, and where marketing campaigns are big enough that they constitute artistic statements in themselves. (Barbie, I’m looking at you.)

Unfortunately, the artworks most likely to run into trouble in this viral hellscape are those that explore complicated, incendiary topics such as sex, race, and identity. Another Goodreads drama played out recently over Everything’s Fine, a debut novel written by Cecilia Rabess and published on June 6. Its plot centers on a young, progressive Black woman who falls in love with a conservative white man in the lead-up to Donald Trump’s election. “It obviously tackles some lightning-rod issues about race, class, and politics and identity in America,” Rabess told me, and so she expected strong reactions on Goodreads and similar sites. “But I think people certainly hadn’t read the book. And so I don’t know how they came to the conclusions that they did—that the book didn’t handle these topics carefully or thoughtfully or intentionally.”

Chalk that characterization up as writerly understatement. “It’s not enemies to lovers if you use it to excuse racists,” a typical one-star review read, referencing a common romance-novel trope. “Some authors shouldn’t be authors bc wtf is this!?” another offered. “i haven’t read this book nor do I plan to but having read the synopsis, I’m rating it 1-star,” a third confessed.

In the case of Everything’s Fine, the pile-on appears to have started on TikTok, where a handful of prominent creators criticized the book. The swell of anger then migrated to Goodreads, where those creators’ fans could register their disapproval. “i didnt and will not even read this i came from tiktok to say i hope the sales are so bad the bookstores have to throw away all inventory because it refuses to sell. anyone who gets an ARC of this should be ashamed,” noted another one-star review.

For Rabess, the experience was brutal. “As an artist, you’re prepared for people to not resonate with the work,” she said. “But I think it feels different when people decide that you yourself are problematic, or you yourself are causing harm, or whatever language they use to describe it. It feels a little bit surreal.” The backlash might have flourished on Goodreads, but it soon escaped to the wider internet. Rabess, who is Black, received angry direct messages and emails, as well as abusive comments under any social-media posts she made. “They said nasty things about me, about my children. Called me coon, other really unpleasant slurs. Told me that I’d be better off dead.”

The anger was scattershot. The commenters using racial slurs clearly knew Rabess’s race, but she wondered if some other online critics assumed that she was a white author intruding on territory they felt should be reserved for writers of color. While authors are sensibly told not to read the reviews—and certainly not to engage with critics—that’s harder when the critics come right up in your (virtual) face and shout their opinions at you.

[Read: The companies that are killing creativity]

As it happens, the podcaster Traci Thomas was among those who disliked Rabess’s book—albeit after reading an advance copy, back in January. “It’s an icky book,” she told me. She objected to what she saw as the moral of the story: Love conquers all, even being a Trump supporter. “The boyfriend in the book, Josh, is wearing a MAGA hat and, like, saying racist shit to [the female protagonist]. And she’s like, It’s fine. And the big revelation for her is that she can still choose to love him. And I’m just like: Okay, cool, go off—and I’m gonna tear this book to shreds.”

Ultimately, Thomas concluded, “I don’t know that the book needs to exist.”

Despite her own strong feelings, Thomas told me that she sometimes felt uneasy about her own reviews being surrounded by knee-jerk reactions and “performative allyship,” even by people whose politics she shared. “There are people who are new to anti-racism work or supporting LGBTQ people, or disability activism or whatever. And they feel it is their job to call out things that they notice without perhaps understanding the bigger historical context.” To illustrate the point, she gave an example: Imagine an author writes a book about Black children riding tricycles, “and then I’ll see a review that’s like, ‘This book didn’t talk about Black preschoolers who ride bikes, and they’re also at risk.’”

That dynamic explains one of the most initially counterintuitive aspects of viral pile-ons: that many seem to target authors who would agree with their critics on 99 percent of their politics. A strange kind of progressive one-upmanship is at work here: Anyone can condemn Ann Coulter’s latest book, but pointing out the flaws in a feminist or anti-racist book, or a novel by a Black female author, establishes the critic as the occupant of a higher moral plane.

The net effect of this is to hobble books by progressive authors such as Gilbert, and by writers of color such as Rabess. The latter is philosophical about the controversy over Everything’s Fine, seeing the backlash as representative of the political moment she was exploring in the novel—of “people feeling a dearth of community and connection, and just wanting a way to connect, a way to express themselves or express their anger.”

Of course, if Goodreads wanted to, it could fix the review-bombing problem overnight. When services that rely on user-generated content are only lightly moderated, it’s always a conscious decision, and usually a cold commercial one. After Gilbert pulled her novel from publication, The Washington Post observed that Amazon, which reportedly paid $150 million for Goodreads, now shows little interest in maintaining or updating the site. Big changes to a heavily trafficked site can be costly and risk annoying the existing user base: Reddit has recently faced down a moderators’ revolt for changes to how developers can access its tools, and Elon Musk’s tenure at Twitter—or whatever it’s now called—will one day be taught at business schools on a slide headlined “How to Lose Advertisers and Alienate People.” A purge of duplicate accounts might sweep up some fanatically devoted Goodreads users—people who can’t bear to share their opinion only once—and make the site feel less busy and exciting.

Goodreads spokesperson Suzanne Skyvara told me by email that the site “takes the responsibility of maintaining the authenticity and integrity of ratings and protecting our community of readers and authors very seriously.” She added that Goodreads is working to “stay ahead of content and accounts that violate our reviews or community guidelines” and has “increased the number of ways members can flag content to us.”

The main Amazon site has several measures in place to stop review-bombing: Reviews from verified purchasers of books are flagged as such to bolster their credibility, while the star rating is the product of a complicated algorithm rather than simply an average of all the review scores. Goodreads could adopt even more stringent measures—but then, it isn’t in the company’s interests to reduce volume in favor of quality, because its entire appeal is based around being a grassroots voice. “Goodreads really needs a mechanism for stopping one-star attacks on writers,” the writer Roxane Gay tweeted after Gilbert’s statement in June. “It undermines what little credibility they have left.” Traci Thomas agrees. In an email, she told me that she would like to see “verified users or reviews that get a check (or something) in exchange for proving they’ve read the book.”

If Amazon will not put the resources into controlling the wrath of Goodreads, then what fairness requires here is a strong taboo: Do not review a book you haven’t read. We should stigmatize uninformed opinions the way we stigmatize clipping your nails on public transport, talking with your mouth full, or claiming that your peacock is a service animal. A little self-control from the rest of us will make it easier for writers to approach incendiary topics, safe in the knowledge that they will be criticized only for things they’ve actually done.

If, as seems likely, Donald Trump is the Republican presidential nominee next year, the 2024 elections will be a referendum on several crucial issues: the prospect of authoritarianism in America, the continuation of a vibrant democracy, the relationship between the executive branch and the other two branches of government, and much else of grave significance.

It will also be a referendum on whether Trump will ever be held legally accountable for his actions. Trump faces multiple civil suits and at least two criminal indictments (with two more seemingly just over the horizon). If Trump were to win reelection, it is almost certain that none of these will ever be resolved—at least not in a way that is adverse to his interests, which any reasonable system would admit as a possibility. Such is the power of the presidency.

Take the simplest cases first: his pending indictment in Florida for violations of the Espionage Act relating to the retention of classified documents, and his anticipated indictment for actions relating to the January 6 insurrection. If Trump is elected, neither of these cases will ever result in a final judgment of Trump’s guilt (much less incarceration).

The Mar-a-Lago documents case is set to go to trial in May 2024. Although many people suspect, not without foundation, that this trial date will slip, let us indulge the possibility that the trial occurs as scheduled and that sometime in June or July 2024, a jury convicts Trump. There is virtually no doubt that Trump would appeal such a verdict, and there is likewise almost no doubt that the presiding judge, Aileen Cannon, would allow him to remain free on appeal.

I have handled several criminal appeals in my career. None has ever been resolved, in even the simplest of cases, in less than a year. That’s just the way our appellate system works; there is no sense of urgency to the proceedings at all (nor, to be clear, should there be in the normal course of business—the trial has concluded and the record is complete; a mature and thoughtful review of those proceedings requires time). The result is that Trump’s appeal of his federal conviction (assuming one is secured) will quite likely still be pending before the Eleventh Circuit, the court with appellate jurisdiction over federal trials in Florida, at least through mid-2025—well past the election and into the next presidential term. And even were an appeal to be concluded quickly, Trump would inevitably then petition the Supreme Court for review, taking yet more time.

A similar calendar will apply to the possible January 6–related charges that may soon be brought by the special counsel in Washington, D.C. Even if that trial were to occur more quickly than the one in Florida (say, for example, in March or April of next year—a big if, given that the federal courts would have to negotiate an efficient trial that does not conflict with the one scheduled for March in New York), the chances of an appeal to the D.C. Circuit, which has jurisdiction over the federal court in the nation’s capital, and thence to the Supreme Court in less than 10 months is near zero.

In short, it seems to me that no possibility exists that any of the federal charges against Trump will be final before January 20, 2025—none at all. And it seems equally certain that one of the very first acts of the Trump-appointed attorney general (whoever that may be) would be for the DOJ to move to dismiss the case or cases against the president at whatever stage they are then pending. Put simply, if Trump wins reelection in November 2024, the federal cases against him will likely be terminated, without final resolution, within 24 hours of his inauguration. That doesn’t mean these proceedings will have been worthless. If Trump has been convicted in either trial, America will have the benefit of a historical record that determines his criminality. But that will be little comfort as we endure another four years of his rule, and as he continues to avoid any semblance of actual accountability.

The situation is more complex when we turn to the state charges Trump faces in a case already pending in New York and another anticipated shortly in Georgia. By definition (at least insofar as the Constitution is concerned), those states are separate sovereigns, and the federal government under Trump cannot direct that those cases against him be dismissed—nor could Trump pardon himself for his state crimes, because his pardon power is, likewise, limited to federal matters. So those cases will proceed.

But boy will they be difficult to bring to resolution.  

To begin with, we can count on the Trump-led DOJ arguing that a sitting president is immune from prosecution by a state, at least during his time in office. And their claim will have some merit. After all, if the New York and Georgia district attorneys can try Trump while he is in office, the prospect exists that any elected official in a state opposed to the president might use his or her power to go after the president on local criminal charges. What’s to stop an elected Republican prosecutor in a very red state from bringing bogus charges against President Joe Biden?

The risk is more than hypothetical. We have already seen how elected attorneys general are using their powers in ever more politicized ways. The leap from “justified” prosecutions to “unjustified” ones lies mostly in the eyes of the beholder. That’s why, more than 50 years ago, the Watergate special prosecutor’s office actually sided with the president on this score, stating that “considerations of federalism would bar his indictment in state court.” Nothing in the text of the Constitution prohibits state prosecution of the chief executive, but nothing authorizes it either, so the question has never been definitively resolved. But if Trump is elected, we can be sure that it will be—and what this Supreme Court would decide is anyone’s guess.

Nor is that the only legal hurdle that the state prosecutors will need to overcome.  Trump’s efforts to have his New York prosecution moved to federal court have thus far been rejected, as have the federal government’s efforts to replace Trump in some of the civil suits against him. Those arguments will, however, have substantially greater force if Trump returns to office; his status as a federal official and the disruption to governmental activity that would arise from his personal liability to civil suit would become significantly more palpable. The Trump-appointed AG would be all but sure to press them in court to the maximum extent possible.

Other challenges may be less legal and more practical in nature. Were New York and Georgia to persist in their cases, the nature of Trump’s retaliation would be limited only by his imagination. What, for example, would happen if he tried to pull federal-law-enforcement funding from those two states? What if he directed the FBI to withdraw from cooperative investigative efforts? What if, in Republican-led Georgia, he pressured the state legislature to pass laws limiting the power of the Atlanta DA or requiring her to dismiss the case? The country should not have to answer these questions.

The prospect that Trump will almost certainly avoid accountability for his criminal conduct if he is reelected is just a small subset of the broader threat he poses to the rule of law. But it is an emblematic possibility redolent with the odor of kingly prerogative. Sadly, the reality is clear: When Americans go to the polls in 2024, if Trump is a candidate, they will not simply be choosing between two political alternatives; they will also be making one of the most important choices in the history of the country. They will be choosing between the modern conviction that no man is above the law and a return to a time when political leaders could act with impunity. Our own national character rests on what choice we make.

In the 2000s, male artists routinely excavated the popular culture of their boyhood for imaginative repurposing in their art. Michael Chabon’s novel The Amazing Adventures of Kavalier & Clay traces the lives of two men who become comic-book creators. In Jonathan Lethem’s The Fortress of Solitude, two boys find a magic ring they use to take on superpowers; the title itself evokes the fictional fortress of Superman. Back then, I remember feeling that the equivalent was not possible for women artists, that the popular culture of American girlhood (horses, dolls, gymnastics) was still considered silly, juvenile—impossible to recuperate as adult art worth taking seriously.

Greta Gerwig seems at last a counterexample. Her entire career as a filmmaker has, in a sense, been a campaign to make art of girlhood materials. Her 2019 film, Little Women, remakes the 19th-century girlhood classic, rendering it freshly urgent for a 21st-century audience. (“I’ve been angry every day of my life,” Marmee, the saintly mother of the novel, says in Gerwig’s version.) Her directorial debut, Lady Bird (2017), captures the angst of a 17-year-old girl’s coming-of-age in all its granular reality, taking seriously adolescent longing and anger.

And now here is the Barbie juggernaut. Barbie dominated the box office on its opening weekend, grossing a stunning $155 million: a post-pandemic high, and the biggest opening for any feature film directed by a woman. Grown-up art about girlhood really can be an economic and cultural powerhouse. In the past week, women I barely know and women I have known for years have texted me their excitement about the film—and then shared their response, which has been in many cases the same: They loved watching the film, and yet more than one felt oddly disappointed by it.

[Read: Greta Gerwig’s lessons from Barbie land]

Perhaps this is because, as the critic Dana Stevens noted in Slate, Barbie “never completely resolves the paradox at its heart”: The movie purports to be critical of Mattel while being enthusiastically sponsored by Mattel. The constraint is so obvious that Gerwig tries to turn it into a joke. The plot hinges on the conceit that all the Barbies mistakenly believe that their existence has turned the “real” world into a feminist paradise. But Gerwig knows that no one wants to watch a hectoring film about how the doll is bad for girls’ self-image. So she subversively jokes about the insidious side of corporate culture while visually celebrating the Barbieness of Barbie. In this way, Barbie really is, if not a wholly successful film, certainly a powerful monument to the strangeness of American girl culture, which remains impoverished in its vision of what young girls are and want.

At once a commercial for a toy and a delicious romp through the pleasures of childhood play, Barbie sits uneasily at the intersection of reality, fantasy, and critique. It is joyous, winking, frothily cinematic. There are lavish costumes, deftly choreographed dance numbers, and aesthetically sophisticated visual puns. Barbie, played by Margot Robbie, is wryly known as “Stereotypical Barbie”; blond and perfect, she showers and drinks without ever getting wet, walks on her toes, floats down from her bedroom to the street as if held by an invisible girl’s hand. The script is peppered with genuinely incisive and funny skewers of gender stereotypes. A key moment in the plot involves the Barbies allowing the Kens to “sing at them” while playing guitars; the preposition is pure genius.

Like many women before it, Barbie tries to be many things to many people. One trailer’s tag line was “If you love Barbie, this movie is for you. If you hate Barbie, this movie is for you.” The line gets at the challenge here, adjacent to the one Stevens identified: How do you make a film for an audience that remains fiercely divided about the material you’re writing about?  

This conflict, one might say, is precisely what forces Stereotypical Barbie out of her dreamy pink home on a quest to understand a series of strange phenomena: cellulite, cold showers, melancholy. We meet Barbie in Barbie Land, where women rule, literally. There is a Barbie president; there are nine Supreme Court Barbies. Kens are reduced to supporting roles. They wait for Barbies to kiss them, notice them, long for them. But when Barbie begins having relentless thoughts of death, she and one of the Kens, played to deliciously banal perfection by Ryan Gosling, enter “reality.” It isn’t the paradise she expected; men ogle her and a girl calls her a “fascist.” As reality’s reality dawns on Barbie, she articulates a set of feelings at once amusingly framed and horrifying to witness: “I feel … ill at ease,” she says. “I’m getting undertones of violence.” (Cue women squirming in the audience.) Meanwhile, Ken is radicalized by his experience: He likes the attention he gets. He sees a new world, one where men are not second-class citizens, and returns to Barbie Land to brainwash the Barbies into serving the Kens. The former Supreme Court justices are now cheerleaders.

Rather than get angry at what has happened to Barbie Land, Barbie is crushed. She stops wearing makeup and lies on the floor in despair, waiting for another Barbie to take the lead in fixing her society. And so the second half of the film tells the somewhat incoherent story of her evolution from thinking she is not “good enough” to becoming a woman bent on self-determination. In the real world, she befriends a real woman named Gloria (America Ferrara), who delivers an impassioned motivational speech about how hard it is to be a woman in a world that demands contradictory things: to care tirelessly for your children but not talk about them too much; to act when action seems fruitless.

The challenge that Barbie, as a cultural object, presents Gerwig is how to mine the performative pleasures of clothes and dance numbers alongside the travesties of modern American culture while framing the story as if the two are wholly disconnected. She’s trying to show what it might really be like to go from a dream land to contemporary America, where Roe v. Wade has been overturned, parents still don’t automatically get parental leave after the birth of a child, and most CEOs of corporations (including Mattel) are still men.

Barbie is at once radical and conventional. It is radical in its light indictment of the patriarchy: One of the very best lines of cultural criticism I’ve heard in ages comes when Ken says, “To be honest, when I found out the patriarchy wasn’t about horses I lost interest.” (The movie might be best read as a brilliant portrait of aggrieved far-right masculinity.) It is somewhat more conventional in its vision of a dream girlhood and its idea of what Barbie is. And perhaps here is where Mattel’s influence is strongest. Despite the presence of Doctor Barbies and Supreme Court Barbies, Gerwig’s Barbies still largely conform to rigid beauty norms; a night of fun is a dance number in glittery and sexy clothes; and no conflict ever occurs, because the days are coated in a vapid gloss of fashion and pink plastic that corporate America has always been selling girls. This is the vision that led my second-wave-feminist mother to ban Barbie from our house. When my aunt gave me two for my fifth birthday, they mysteriously disappeared a few days later.

[Read: The surprising key to understanding the Barbie film]

To some degree, this vapidity is Gerwig’s point: Barbie must leave Barbie Land, as all girls must, to confront a messier and more complex reality. And yet Barbie skirts the deep cognitive dissonance the doll inspires in many girls. Thanks to my mother’s ban, I played with Barbie only at friends’ houses. I recall finding the doll boring, because she was relentlessly tethered to our world and to the culture’s expectations for adult me, ones that didn’t yet interest me (boys, houses, Malibu). Instead of playing Barbie, my friends and I disfigured her so she looked like the film’s Weird Barbie (Kate MacKinnon), a doll with a 1980s punk haircut who’s out of step with all the others. Like the montage in the film showing what a girl did to Weird Barbie, we chopped off her hair down to chunky sprouts, turned her head backward, covered her face in green marker. In the film’s lingo, Weird Barbie is weird because girls have “played too hard” with her, a telling evasion. Weird Barbie is not weird because someone played too hard with her. Weird Barbie is weird, as Gerwig surely knows, because some girl used her to express all her frustration at the confection of materialism and reductive beauty norms. Weird Barbie wants to raze Barbie Land, not just help Stereotypical Barbie see reality for what it is.

Barbie is a fascinating moment in the ascendance of girl culture, a brilliantly twisty dive into the core of the contradictory scripts shaping what it is like to grow up as a girl in America. One of those scripts teaches that girlhood isn’t a space of imaginative autonomy but a kind of prep school for a conventional consumer adulthood that will be shaped by hair appointments, shopping, and a job. Despite the constraints she faces, Gerwig tries to invoke a more capacious script for girlhood at the film’s end, in a beautiful montage of girls playing around the world, in muted visuals that starkly contrast with the Technicolor Barbie world. If the film emerges aesthetically scathed, that doesn’t mean it is a failure. I left delighted by how much Gerwig had achieved through humor, and by the conviction that Barbie can only help make room for more films by women about women. The film’s tag line is right. One can hate Barbie and still be glad this movie was made, because here it is at last: a movie full of jokes for us, the women and girls in the room. In that sense, nothing about Barbie is silly or juvenile.

Through "preemption" measures, ballot and voting restrictions, gerrymandering and other schemes, America continues to be threatened by entrenched minority rule.