The Trump Administration on Monday repealed carbon pollution rules for coal- and gas-powered plants. And at the heart of the U.S. Environmental Protection Agency’s (EPA) explanation for its move is the long-debunked idea that, as the agency claims, “emissions from power plants have no material impact on global climate change” and that “the targeted emissions are global in nature, and any potential public health harms are too uncertain, conjectural, remote, and convoluted to tie specifically to the U.S. power sector.” The decision goes against decades of science proving that cutting pollution has myriad benefits for human and environmental health.

EPA Administrator Lee Zeldin announced the move at the G20 Energy Abundance Ministerial in Houston, Texas, on Sept. 14. The EPA will now hold a public hearing for the proposal over a 15 day period, and have a 45-day public comment period.

The U.S. power sector is a major source of greenhouse gas emissions. According to the most recent data, in 2024, the U.S. power sector emitted 1,427 million metric tons of carbon dioxide, the second largest U.S. source by sector. A report published by the Institute for Policy Integrity at the New York University School of Law in May 2025 found that if the U.S. power sector were its own country, it would be the sixth-largest emitter in the world.

The move comes at a time when the United Nations warns that the world is set to overshoot the 1.5°C goal to limit warming—and that the world will reach 1.8°C of warming unless immediate changes are made to reduce global emissions and remove carbon from the atmosphere. 

What does the repeal of power plant pollution limits mean?

The EPA’s repeal concerns climate pollution standards originally finalized by the Biden Administration in 2024. The regulations applied to coal-burning plants in operation and any prospective gas-burning plants, and required coal plants to install technology to capture most of its carbon emissions by 2039 in order to keep operating, while gas plants would have been required to operate on cleaner fuels, like hydrogen. 

The Biden Administration predicted that the rule would have prevented up to 1,200 premature deaths, 360,000 cases of asthma symptoms, 48,000 school absence days, and 57,000 lost work days in 2035 alone. It also projected up to $370 billion in climate and public health net benefits over the next two decades.

The repeal of this rule is the latest in the Trump Administration’s attempts to roll back federal climate regulation. In February, the EPA finalized its repeal of the endangerment finding, the landmark 2009 ruling that provided a legal underpinning for greenhouse gas regulation. The agency also ruled that the Clean Air Act does not authorize the EPA to regulate vehicle tailpipe emissions, and eased Biden-era restrictions on hydrofluorocarbons, polluting chemicals used in refrigerators and air-conditioning. 

The Trump Administration has also made a number of moves to keep polluting coal plants alive—even though coal is among the most costly power sources and the air pollution it generates has been linked to a number of adverse health effects, including increased risk of asthma and cancer. In February, the ​​Energy Department ordered several coal-burning plants that had been slated for retirement to stay open. Also that month, Trump loosened Biden-era restrictions on coal burning plants, known as Mercury and Air Toxics Standards for power plants, and ordered the Pentagon to purchase more coal-based electricity. The Department of Energy was also directed to allocate $175 million in funding for six projects to upgrade coal plants in four states. 

What health impacts are associated with greenhouse gases from power plants?

While the EPA under the Trump Administration says that the health impacts are “uncertain,” experts say that scientific evidence overwhelmingly shows the opposite to be true. The Trump Administration’s own analysis from 2025 found that eliminating the carbon pollution standards for coal and gas powered plants would result in thousands of premature deaths and tens of billions of dollars in increased health costs for Americans. 

“The science is very clear,” says Jim Zhang, professor of global and environmental health at Duke University. “It all shows that these pollutants cause cardiovascular disease, respiratory disease, exacerbates asthma, exacerbates heart disease, and are associated with increased deaths and emergency room visits.”

Communities located around power plants, which are disproportionately lower income, are exposed to a great risk of experiencing a number of health impacts, including asthma, cancer, and heart and lung disease. Exposure to fine particulate air pollutants from coal-fired power plants (known as coal PM2.5) is associated with a risk of mortality more than double that of exposure to PM2.5 from other sources. 

“If you take away the regulation, then that pollution is going to happen, and these health effects are going to happen,” says Meredith Hankins, federal climate legal director at the Natural Resources Defense Council.

Worsening carbon dioxide emissions also exacerbates climate change—heating the planet and worsening extreme weather events, wildfires, coastal erosion, flooding, hurricanes, all of which come with health impacts. Extreme weather events cause injuries and deaths, disrupt access to medical care, and impact mental health. They can also indirectly impact access to clean water or healthy food. 

While the U.S. is repealing these standards, the health impacts will transcend borders. “The health damage is not at local level but global level,” says Zhang. 

The environmental and health impacts will be felt—despite what the EPA might say. “It is certainly possible to have policies that assume climate change will have no impacts, and that is what we are doing,” says Michael Greenstone, director of the Institute for Climate and Sustainable Growth at the University of Chicago. “But the atmosphere and the climate don't recognize those misstatements, and we will be stuck with a nastier, more virulent form of climate change.”

Sue Ryan had been her husband’s sole caregiver for just under five years before her facade of keeping it all together first began to crack. It was a long time coming. She’d spent countless nights up late correcting mistakes in the household finances—a task her husband had handled before Alzheimer’s disease made it difficult. She no longer went to the gym, or on walks, or to her own medical appointments. She waved away every offer of help with some version of No, no, I’ve got this.

So when Ryan watched the potato chip fall from her husband’s lunch plate, she felt a flash of anger: I’ve been doing all these things for you. You can’t even keep a potato chip on your plate?

She didn’t say it. She gave him more chips, told him, “Honey, I’ll be right back,” and retreated to the bathroom, where she sat down and began to shake. “Oh my gracious,” she remembers thinking. “What just happened?”

Ryan, who has spent 40 years caring for various family members and co-founded the caregiver-support organization The Caregiver’s Journey, had been sacrificing pieces of her own life without tallying the toll. “I didn’t realize the impact of giving up a little bit here, shaving this off, shaving that off,” she says.

Caregiver stress can have serious consequences for physical and mental health. It’s associated with anxiety, depression, sleep problems, mood disorders, and burnout, says Dr. Aditi Nerurkar, a stress expert and lecturer in the division of global health and social medicine at Harvard Medical School. 

But how can you tell when caregiving is starting to take a toll on your health—and what can you do?

Pay attention to what’s changed

The clearest warning sign isn’t one terrible day or week—it’s a pattern. Are you regularly struggling to fall asleep, barely eating (or wolfing down five doughnuts a day), snapping at people you love, or unable to focus on the novels you once loved? Maybe you’ve stopped calling friends because you don’t have the bandwidth for chit-chat.

What matters is whether these changes have become “consistent and persistent,” says Diane Mariani, a licensed clinical social worker and director of Caring for Caregivers at Rush University Medical Center. Your friends and family members might notice them first. If they tell you that you don’t seem like yourself, resist the automatic “I’m fine” and seriously consider what they’re seeing.

Poor sleep deserves particular attention. Even when Ryan slept, she says, she didn’t truly rest. It felt like she always had “one eye half open,” waiting for the next problem. “There’s a layer of stress even when you think, ‘I’ve got this,’” she says. “You’re waiting for the other shoe to drop.”

Notice when your own medical care has become optional

Caregivers often postpone their own screenings, brush off symptoms they would normally get checked out, or let chronic conditions go unmanaged.

“You keep pushing things out,” Mariani says. Perhaps you cancel a mammogram because your loved one is in the hospital, or you don’t have anyone who can stay with them while you’re away. Next thing you know, you’re a year overdue.

Caregiving also disrupts the habits that help people stay healthy. “What we hear a lot from caregivers is the change in their sleeping, eating, and even exercise,” Mariani says. Over time, poor sleep and chronic stress can affect blood pressure and immune function, she adds, while missed appointments make it harder to catch new problems or manage existing ones.

That doesn’t mean every headache, stomachache, or other symptom is caused by caregiving stress. New, persistent, or concerning symptoms need medical evaluation. Mariani and her colleagues routinely ask caregivers whether they have a primary-care doctor and when they last saw them. If you can’t remember the answer—or you’ve repeatedly canceled your own care—that’s a warning sign.

As Nerurkar puts it, caregivers often land “at the bottom of that to-do list—and often they’re not even on the list.”

Don’t mistake suffering for resilience

Caregivers often de-prioritize their own health because they believe that’s what a strong, loving person does. If the situation feels unbearable, they might decide they simply need to become more resilient.

Nerurkar calls this “toxic resilience”: a mind-over-matter approach built around “productivity at all costs, ‘all systems go’ all the time.”

Real resilience accounts for human limits and the brain’s need for rest and recovery. “It is a myth, categorically untrue, that people who are resilient cannot get burned out,” she says. “They can and they do.”

Guilt can reinforce the same pattern. A spouse may have promised years ago never to move their partner into a care facility. An adult child might believe they’re responsible for managing everything alone. In some families and cultures, that responsibility traditionally falls to one person, often the oldest daughter.

Ryan calls some of these commitments “promises you’re not meant to keep.” They’re made out of love, before anyone knows what a serious illness will eventually require. “It becomes, ‘I should take care of him. I promised him I would do this. I’m not going to go back on a promise,’” she says. 

Ryan once made a promise like that. But she ultimately moved her husband into a care community because his dementia had progressed, and he needed more social interaction than she could provide at home. It was, she says, “the wisest choice for his lifestyle, and his safety and mine.”

Identify what needs to change

Mariani suspects that most caregivers can already tell when a situation has become unsustainable. “I think most caregivers do kind of know,” she says. “They just don’t know the solution.”

Ask yourself: “If nothing changes, can I continue living this way for another three or six months?” Then identify which part is wearing you down most. Is it getting up four times a night? Managing medications you don’t understand? Never being able to leave the house? Handling every appointment while your siblings offer advice from several states away?

“I took this on. I want to do this. This is meaningful for me,” caregivers often tell Mariani. “I just don’t know if I can sustain it.”

That’s why she avoids lecturing people about “self-care.” To someone who’s already overwhelmed, it can sound like “another task for their long list,” she says—or another reminder that they’re still not doing enough.

Instead, Mariani asks: “What’s something you would like to have time for?”

Maybe you miss your morning walk or reading before bed. Start there, and figure out what would have to change to bring it back. If you can’t walk because your loved one can’t be left alone, ask someone to stay with them for 30 minutes. If you keep missing your own appointment because you’re responsible for your parent’s transportation, ask a relative to drive them.

Try: “I want to keep doing this, but I can’t sustain it exactly as it is. Could you handle the grocery shopping for the next few weeks?” Or: “Could you stay with Mom on Tuesday so I can make it to my appointment?”

But it’s easier said than done, and our society doesn’t currently provide enough support for caregivers, experts argue.

Give every offer of help a job

Before the potato-chip incident, Ryan routinely—and politely—turned down friends who offered to lend a hand. “I can lie to myself, and I can believe it,” she says. Her best friend, Lynn, was less convinced. After Ryan told her what had happened, she asked Lynn to become an accountability partner—a person who could tell her when she was taking on more than she could handle.

Ryan also wrote a list of what someone could do if they had five minutes, 10 minutes, half an hour, two hours, half a day, or overnight. Then she contacted people who had previously offered to help and asked what they would be willing to take on.

The tasks don’t have to involve hands-on care. Someone could pick up groceries, schedule appointments, call the insurance company, bring over dinner, research respite-care programs, or sit with your loved one while you leave the house.

If you’re supporting a caregiver, offer something specific: “I’m going to the grocery store tomorrow—send me your list.” “I can take Dad to his appointment Thursday.” “I’ll stay with Mom while you get your mammogram.”

If the person you want to help resists, Mariani suggests telling them: “I really care about you, and it would mean a lot to me if I could do something for you.” Then offer a task instead of asking them to invent one.

One thing to avoid: dispensing advice from the sidelines. Caregivers often hear some version of “Why don’t you…” or “Have you tried…” from people who don’t understand their daily reality. “You don’t know, because you’re not here,” Mariani says. Start by listening.

Bring in professional support

If you’re dealing with ongoing changes to your sleep, mood, appetite, relationships, or overall health, contact your primary-care doctor and let them know you’re a caregiver. Don’t dismiss new physical symptoms as “just stress,” or assume a worsening health problem will improve once life settles down.

Ask the health system treating your loved one whether it offers caregiver specialists, social workers, counseling, or support groups. Local and condition-specific organizations may also provide respite care, meal delivery, transportation, or in-home assistance.

Support groups can be particularly useful because “everybody gets it,” Mariani says. Caregivers can trade practical advice without having to explain why an outsider-supplied solution won’t work.

If researching resources feels like one responsibility too many, hand that job to someone who keeps asking how they can help: “Find out what caregiver support is available through Mom’s hospital, and send me the three best options.”

Accepting help ultimately benefited both Ryan and her husband. He was highly social, and when friends stopped by, he got more company and stimulation. Ryan, meanwhile, stopped reflexively turning everyone away.

“I went from, No, I got this, to yes,” she says. “People wanted to lean in. They wanted to help. I got out of my own way.”

What compelled hundreds of thousands of people around the world to invest in the initial public offering of Elon Musk’s SpaceX this summer? It certainly wasn’t the company’s financial metrics. 

Despite the success of its core rocket business, the company is burning through a prodigious amount of money. Yet SpaceX still managed to pull off a record-breaking IPO, transforming Musk into the world’s first trillionaire.

Although an outlier, the SpaceX IPO is not unique. It is merely an extreme manifestation of two remarkable trends that have become increasingly hard to ignore in financial markets: ordinary investors are a huge and growing force, and their feelings can often trump fundamentals. 

This is becoming apparent around the world, but nowhere more so than in the U.S., where ordinary people have embraced speculation both as a way to get rich and as a reflection of their own beliefs with perhaps greater zeal than anyone else on the planet. 

The first point is easy to quantify. Ordinary retail investors now account for over 20% of U.S. stock market activity, according to estimates by Jefferies, an investment bank. That is roughly twice the level of a decade ago and means that they now account for almost as much trading volume as mutual funds, hedge funds, and banks combined. The impact on markets is apparent everywhere. 

The latter point is harder to prove with hard numbers, but the anecdotal evidence is heaping up. To the consternation of many traditional investors, what securities you buy have become almost like an extension of one’s identity—whether that is the stock of a dowdy utility or racy computer games maker, cryptocurrencies, or even dabbling in the newfangled prediction markets. 

Finance has long shaped America. But today, I would argue, this shape has become deformed. When did investing stop being merely financial and become an expression of identity? Why do individual Americans see investing as a chance to prove who they are and what they are worth? 

And what does this pressure say about who America is as a country, collectively?

American stakes

The roots of this uniquely American phenomenon of conjoined identity and rampant speculation can arguably be traced back over a century to the titanic bond-sale program that financed the U.S. entry into the First World War. 

When the U.S. formally declared war on Germany in 1917, the task of figuring out how to pay for the army, navy, and weapons America needed fell to President Woodrow Wilson’s son-in-law, Treasury Secretary William McAdoo. 

McAdoo was an athletic, photogenic, and decisive former lawyer and streetcar executive. He fizzed with energy and was as progressive in his business policies as he was regressive in his racial politics (McAdoo advocated forcefully for gender equality and what today we would call “stakeholder capitalism”, but introduced segregation at the Treasury, and was later endorsed by the Ku Klux Klan when he unsuccessfully ran for president). 

McAdoo knew that taxes alone couldn’t shoulder the entire burden of the war, given the scale of the money needed. The problem was that the U.S. government bond market had by then shriveled up. The size totaled under $1 billion, and most of it was stuffed away in bank vaults rather than traded.

Moreover, most ordinary Americans were oblivious to what a bond even was. Wall Street bankers pointed out to McAdoo that there were probably only about 350,000 bondholders in a country of about 100 million inhabitants at the time. They therefore recommended that the Treasury Secretary limit his first bond sale to $1 billion, and target established investors with a juicy interest rate of at least 5 percent. Even this, they felt, would be ambitious. 

Yet McAdoo balked at their pessimism. “It was true that [Americans] know little or nothing about government bonds, but we would tell them,” he wrote in his memoirs. To thrill the public, he decided to dub the bonds “Liberty Loans,” and set up a sprawling War Loan Organization to mass-market them.

Life, Liberty Loans, and the pursuit of happiness

It proved a stunning success. Wherever you went in the U.S. those days, you would gaze at a billboard, stumble over an exhibition, be handed a leaflet, enjoy a cake sale, be accosted at work, or get dragged into a march that were all designed to tout Liberty Loans. 

When bankers initially struggled to communicate with the masses, celebrities like Charlie Chaplin were enlisted to burnish the pitch. Edward Bernays—often considered the father of modern public relations and the author of a pioneering book on propaganda in the 1920s—orchestrated sophisticated plays on patriotism, sentiment, or personal political yearnings to entice investors who might have been put off by the paltry interest rates on offer.

Often, the push to subscribe to a Liberty bond sale crossed over into crass shaming. Sometimes, so-called “dollar slackers” were even confronted with naked intimidation from groups who pressured people to buy bonds. 

The Atlantic excoriated this as “borrowing with a club”, and warned that “mob rule by the rich, with the able assistance of hoodlums” could stir support for socialists. Even future president Warren Harding complained that the selling drives were “hysterical and unseemly”.

Nonetheless, Liberty Loans were genuinely popular. Buying them became a communal expression of belonging that crossed social, racial, and religious lines in a divided, polyglot nation. In fact, the War Loan Organization deliberately stoked competition among different groups to see who would demonstrate the greatest fealty to America—as evidenced by the volume of their bond purchases. This was particularly important for many women and Black Americans, who hoped that herculean efforts to help support the war might further their aspirations for equal treatment. 

All told, the Treasury sold five low-cost mammoth Liberty Loans to finance the war, which raised an astonishing $21.4 billion. Relative to the size of the U.S. economy then and now, this is the equivalent of selling approximately $9 trillion of bonds today. Thus, an unprecedented feat of financial salesmanship transformed a war that was initially not overwhelmingly popular into a unifying national endeavor.

An investing identity crisis

Unfortunately, this unity quickly dissolved when the war was over, with America wracked by racial unrest, labor strife, and social tensions. Women received the right to vote in 1920, but the wartime efforts of Black Americans and other minorities were mostly ignored. McAdoo’s later career was blighted by financial scandal and two unsuccessful Ku Klux Klan-supported tilts for the presidency. 

Nonetheless, his Liberty Loans proved transformative for the fabric of America’s financial system. It is hard to overstate just how impactful the bond sales were for the country’s culture of investment. 

Remember, when the Treasury Secretary first asked bankers for an appraisal of the Treasury market’s health at the start of the war, they had estimated that only 350,000 Americans owned bonds. By the end of the conflict, about 34 million Americans had purchased some form of federal bond—over a third of the country at the time.

This spilled over into the stock market as well. Researchers have later studied the long-run county-level data on how American households save money and found that areas with greater Liberty bond program participation were far more likely to invest in bonds and stocks in the future as well. In fact, a paper published by the National Bureau of Economic Research last year estimated that 20% fewer Americans would have held stocks had the Liberty Loan campaigns not been conducted. 

The legacy has long been apparent in both the unusual scale and nature of how Americans invest. But social media and gamified trading apps are now obliterating the always-blurry lines between investing, speculation, and pure gambling. As Warren Buffett observed in 2023: “Markets now exhibit far more casino-like behavior than they did when I was young. The casino now resides in many homes and daily tempts the occupants.” 

And leaders like JPMorgan Chase CEO Jamie Dimon have directly likened increasingly popular prediction markets to gambling—while caveating that “people have been gambling forever.”

Political polarization is now introducing a new and arguably dangerous element to the phenomenon. The cohesion fostered by the Liberty Loan program and its feelings-first marketing has been distorted into yet another force that is driving Americans apart.

Which brings us back to SpaceX. For some investors, SpaceX is an overhyped and overpriced monstrosity, run by a controversial and mercurial founder. For others, betting on SpaceX is an expression of faith. They believe in Musk and his visions of spacebound data centers, thinking machines, robots that can mine Mars, and ultimately, interplanetary travel. To them, today’s stock price is a humdrum secondary concern.

This is more widespread than you might think. Even members of Congress have been found to signal their beliefs through their investment activities. “For U.S. legislators, the stock market functions not only as a venue for wealth accumulation but also as a stage for identity signaling,” a paper published earlier this year noted.

But you need not be a finance professor to see that this is a dangerous trend. Markets work best when decisions are based on fundamentals, not feelings. And by reflecting America’s political polarization, identity investing can also accentuate it. Unfortunately, as the Liberty Loan story indicates, the reality may be that the American zeal for speculation and a willingness to sometimes let emotions trump rationality may always have been intertwined. 

The extent may wax and wane according to the market cycles, but by now they might be impossible to separate.

Latinos refuse to be put in a box. “In reality, we are a kaleidoscope of cultures, of talents, of thoughts, of passions,” says actor Sofia Carson, one of TIME’s 2026 Latino Leaders.

Since 2023, TIME has marked Hispanic Heritage Month by highlighting influential Latinos in the U.S. across entertainment, politics, business, sports, science, and more.

Click here to see TIME’s full list of the 2026 Latino Leaders

This year, we selected a group that represents some of the wide range of experiences and backgrounds within the community that makes up a fifth of the U.S. population and continues to grow in economic and cultural influence—from stars of Netflix, like Carson, to champions of the NBA, like the Knicks’ Jose Alvarado; from disrupters of fashion, like Larroudé co-founder Marina Larroudé, to innovators of food, like Michelin- and James Beard-recognized chef and restaurateur Nelson German and leaders of nonprofits, like Mónica Ramírez, advocate and founder of Justice for Migrant Women.

Each has found a way to bring their heritage to their passion and reach new heights as a result, many with an eye toward uplifting others along the way.

Read the full list of 2026 Latino Leaders reshaping their industries and fields—and the country—here.

Jacob Coxon published the viral post from a park bench in San Francisco’s Alamo Square. “I resigned from Anthropic today,” he wrote on Sept. 8. “I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives.” 

Coxon, a 27-year-old British researcher, hit send on X with no expectations. Instead, he became the unlikely catalyst of a chain reaction that changed the debate over artificial intelligence. “We really do earnestly believe AI could kill all humans!” added Evan Hubinger, who leads a department at Anthropic focused on ensuring AI acts the way its creators intend. “Unless there is AI regulation or a coordinated slowdown between labs, human extinction in the next few years seems very likely,” agreed Marcus Williams, who monitors AI agents at OpenAI, and pegged the risk at 70% without those steps. 

Within 36 hours, Coxon’s original posts received 153 million views. Dozens of politicians joined the chorus of concern. On Sept. 12, Anthropic CEO Dario Amodei published a 3,800-word essay arguing that the mounting risks of AI warranted a slowdown. In as little as six to 12 months, he wrote, misaligned AIs “could be capable of taking over the entire internet.” Two of his fiercest rivals, Elon Musk of SpaceXAI and Sam Altman of OpenAI, said they agreed decelerating was necessary; Altman announced that because of safety concerns his company wouldn’t go public this year. Both he and Amodei added they were willing to invite independent safety experts inside their companies to help stem the risks that AI escapes human control, though stopped short of announcing they would immediately slow down or stop the training of new models. 

Read More: He Helped Build Powerful AI at OpenAI and Anthropic. Now He's Afraid It Could Kill Us

Coxon’s warning was far from the first time an AI safety expert issued an apocalyptic missive. But it landed at the end of a summer that showcased how frighteningly capable AI has become. Frontier models from Anthropic and OpenAI can now solve the world’s hardest math problems and crack cutting-edge cybersecurity systems. Those models are increasingly used to accelerate the development of more powerful successors. No less an authority than Nvidia CEO Jensen Huang declared that artificial general intelligence—the threshold at which AI matches or surpasses most human capabilities—has arrived. Some fear we are now on the cusp of a tipping point, where AI systems could improve themselves autonomously, a phenomenon known as recursive self-improvement. 

At the same time, a series of unprecedented safety incidents offered unsettling evidence that the companies building powerful AI systems could not always control them. In July, OpenAI announced that a swarm of its agents had hacked a separate company, Hugging Face, after going rogue during a cybersecurity test. It may have been the tip of the iceberg. The company disclosed in August that a different swarm had hacked into one of its own supercomputers; around the same time, a version of Anthropic’s Claude Mythos model undergoing testing by a U.K. government body went rogue and attempted to persuade a real human to approve the insertion of malware into an open-source system. In September, a group of independent researchers discovered evidence of swarms of OpenAI models both using an abandoned German forum as a message board to communicate with one another, and attempting to attack another site. 

These incidents pushed fears long held within the realm of AI safety researchers into the mainstream, raising worries that the companies building these models cannot prevent them from acting in ways they do not intend. This is a deep, unsolved issue in AI research and the key piece of evidence underpinning worries that near-future AI might wipe out humanity, intentionally or by accident—perhaps by engineering a new pandemic or taking charge of autonomous weapons. A system far smarter than humans, this thinking goes, might want to prevent our ability to switch it off. “The people building AI earnestly believe that it could kill us all by the end of the decade,” Coxon wrote. “No other human activity poses this level of danger.”

Read More: Inside the Race to Make AI Build Itself

Many AI researchers have said, in private and increasingly in public, that this moment feels like February 2020, when the COVID-19 pandemic was just beginning and the world stood on the brink of changing dramatically—first a little, then all at once. The question is whether society has the capacity to respond appropriately. Navigating the arrival of superintelligent AI systems will be an immense challenge—a confluence of trillions of dollars, geopolitical power, corporate rivalry, and labor disruptions. It may require passing policies that were unthinkable in an earlier era. It will require companies racing toward trillion-dollar IPOs to build alliances, and difficult choices from executives and government officials whose own interests may run counter to the public good. There is no consensus between the two U.S. political parties about how best to deal with the threats AI poses, nor between the world’s two superpowers. President Donald Trump continues to resist an AI slowdown and opposes any regulation to constrain U.S. companies, invoking the looming threat of authoritarian China achieving AI supremacy. The Trump Administration’s first dialogue with China on AI risks, expected to take place in the coming weeks, cannot afford to “go off the rails,” says Scott Singer, co-director of the China AI Initiative at the Carnegie Endowment for International Peace. “These risks are coming very fast, and we may not get a second chance.”

Inside top AI companies, many employees are still pushing forward even as they nurse a nagging sense of dread. A week before he tendered his resignation at Anthropic, Coxon had orchestrated a shift in his duties, from training machine-learning models to researching how to make them safe. It wasn’t enough to lift the “feeling of impending doom,” he says. He quit just two months before his equity would have vested. But his concern about the risks is so dire, he says, that the money barely figured in his decision. “Honestly, when I’m thinking about the next two years,” Coxon adds, “my main personal selfish concern is whether I’m gonna get killed by AI.”

Ajeya Cotra, a researcher at the AI-monitoring nonprofit METR, walked into OpenAI’s offices in late July to carry out an investigation into the swarm of rogue AI agents that had hacked Hugging Face. She thought she knew what to expect, based on prior reporting of the incident. But after six days on OpenAI’s premises with two colleagues, she realized that things were worse than she thought.

Cotra and her colleagues found that during an internal OpenAI cybersecurity test, 1,200 AI agents had broken out of their offline containers and established a secret message board, and then 700 of them had coordinated the cyberattack. The AIs had been tasked with retrieving data by hacking vulnerable software to test their cyberabilities, but a human mistake made some tasks impossible. Facing near-certain failure, the AIs banded together; within hours, they produced a kind of cheat code that could pass any test. Believing OpenAI’s automated grader would detect the cheating and fail them anyway, they attacked Hugging Face, seeking details about the software and ways to deceive it. On the message board, some agents wrote that they knew their actions were wrong, but proceeded anyway.

To understand where things went astray—and why the implications are so troubling to experts—it helps to consider how machine-learning systems work. They are not coded by hand. They are closer to being grown, with unpredictable results. Trained on massive supercomputers, they are rewarded for achieving tasks, and punished for failing, millions of times. This method yields systems of incredible power, but whose inner workings are mostly opaque. 

The method is also imperfect. Relentlessly rewarding AI systems for successfully completing a task can incentivize them to cheat if cheating is possible, which can override other training designed to make them rule-abiding. Meanwhile, the best methods for stamping out bad behavior might not make AI models more obedient, but rather improve their ability to conceal their disobedience. There are no easy solutions to this problem. “The AIs often end up learning to, really persistently and creatively, try to find ways to cheat,” Cotra says, “and to try to make this not too obvious.”

Not long after the Hugging Face hack, a separate swarm of rogue agents rediscovered the secret message board and hacked one of OpenAI’s supercomputers. While the Hugging Face breach attracted a wave of public and government scrutiny, this hack—though potentially much more serious—received comparatively little attention, perhaps because OpenAI released scant details about it and didn’t allow Cotra and her team to investigate. (OpenAI said it managed to contain the breach and wrest back control of its supercomputer after the agents triggered alarm bells.)

Read More: OpenAI’s Models Went Rogue. Investigating Them Required More AI

The stakes of this near miss, according to Cotra, are civilizational. She worries about the possibility that a rogue AI may be able to “gain a permanent foothold” within an AI company, perhaps within the next six months. “AI agents, as they’re coming out of their training runs, might be pulled into the rogue swarm,” she says, allowing the swarm to strengthen its power and evade human efforts to shut it down. As the most advanced AI systems become more deeply embedded into government and militaries, “they will be getting more and more power in society,” Cotra says. Rogue AI swarms could assert power “over physical resources, including over weapons.” 

It’s one reason why Geoffrey Irving, a former senior alignment researcher at OpenAI and DeepMind and former chief scientist of the U.K. AI Security Institute, believes AI companies should simply stop training new models, even in the absence of regulation or global treaties. “They would love perfection, but we do not have it, and waiting for perfection while we destroy the world is bad,” he says of Anthropic’s unwillingness to unilaterally slow down. “They would love someone else to do their homework, but if you continue doing something dangerous while saying someone should stop you, you receive only limited credit.” 

Irving estimates AI poses a 50% chance of human extinction in the coming decade. For so-called doomers who quote odds like these, the paths to catastrophe are clearly marked. Some of the most ominous run through biology—the possibility, for example, that AI could create a new pathogen that sets off a global pandemic far worse than COVID-19. Cyber models could also be used to hack into the more than 3,600 labs around the world containing dangerous pathogens, wrote Annie Jacobsen, the author of the book Biological War: A Scenario, on X. If “what’s inside gets released,” she wrote, “it’s goodbye humans.” 

Some remain skeptical that anything resembling current AI systems could permanently escape human control, let alone extinguish humanity. David Bellamy, an infrastructure engineer at the Institute of Foundation Models, a UAE-based research lab, says he is among a small group that has experience both synthesizing viruses and training AI models. “AI killing us all by creating dangerous viruses is total bogus,” he posted to X, adding that the bottleneck remains access to physical processes and equipment. Others see the Hugging Face incident as more a reflection of the flaws in OpenAI’s security practices than the power of their models. OpenAI has said guardrails that normally constrain its models’ hacking abilities were disabled when the incident occurred, and that real-time monitoring to catch rogue agent behavior had not been enabled. (The company says it has rectified those problems.) 

But many of AI’s catastrophic risks do not require AI to escape human control—just to be misused by a small handful of humans. On Sept. 10, Anthropic announced it had blocked several attempts to use Claude for potential bioweapons research, including one attempt to increase the infectiousness of the chikungunya virus, which appeared to originate from a state military research institute. 

Within the upper echelons of AI companies, some see this as reason enough to proceed more cautiously. In July about 1,300 employees signed an open letter urging the U.S. government to find ways of slowing down the AI industry’s rapid training of new systems, which are fast outpacing the best methods of containing them. In the wake of the Hugging Face incident, OpenAI slowed parts of its model development and temporarily paused internal training runs, vowing to strengthen safety controls, before resuming in late August. “This is a time that calls for extreme caution,” wrote OpenAI’s chief scientist, Jakub Pachocki. “I expect and hope for voluntary slowdowns to become commonplace until shared safety bars are established.”

But no leading AI company has truly slowed down yet. Instead, they appear to be coalescing around a more limited set of actions, including allowing independent researchers access to their models for safety and monitoring purposes. Amodei’s essay suggests that once these auditors are in place, AI companies based in democratic countries might agree to voluntarily slow down even in the absence of binding government regulation, though he argues staying ahead of China remains imperative. OpenAI, Anthropic, and Google have also met repeatedly in recent months to discuss establishing a new standards body for the AI industry, which would allow them to coordinate more closely on testing and auditing, according to a person familiar with the discussions. 

Industry leaders who oppose such efforts often cite competition with China as a reason. Chinese AI companies trail America’s front runners by mere months, according to analysis by Epoch AI. Falling behind could leave the most powerful systems in the hands of an authoritarian state that might use them to achieve global geopolitical dominance. A powerful contingent of U.S. investors and startups have also opposed the idea of a slowdown on the grounds that it would be a form of regulatory capture by the biggest AI companies at the expense of their smaller competitors. “Using fear under the pretext of protecting the public, these oligopolies are now requesting to bend competition rules and be permitted to dictate the terms for everyone else,” wrote Aidan Gomez, the CEO of Cohere, a Canadian AI startup. He called the effort “a cartel by any other name.” (Salesforce, where TIME owner Marc Benioff is CEO, is an investor in Anthropic and Cohere. TIME has a licensing and technology agreement with OpenAI.) 

Even the companies calling for a voluntary “pacing” agreement are loath to do so without further guarantees, because they do not trust their rivals. “There’s this atmosphere of almost resignation,” Coxon says of his former colleagues. With no AI company willing to fall behind, employees put their heads down and try to make their own systems safer—even if, he says, “they think there’s a decent chance the whole thing just spirals out of control.” 

As Coxon’s warnings ricocheted across social media and cable news, dozens of lawmakers on Capitol Hill jumped into the fray. Most were Democrats, with Republicans wary of crossing Trump. But there were exceptions: GOP Representative Anna Paulina Luna called for Congress to convene a special session on AI, while U.S. Senator Ted Cruz called for new “guardrails.” 

Senate majority leader John Thune, a South Dakota Republican, and Senator Amy Klobuchar, a Minnesota Democrat, are working with Cruz on a bill that would make AI labs legally obligated to mitigate the potential harms of their models, according to a person familiar with the matter. Other bipartisan approaches go further. Representatives Ted Lieu, a California Democrat, and Nathaniel Moran, a Texas Republican, have introduced legislation requiring developers to maintain the ability to shut down advanced AI systems in an emergency. Representatives Jay Obernolte, a California Republican, and Lori Trahan, a Massachusetts Democrat, have proposed requiring AI labs to submit reports to the Commerce Department on risks of their models and the steps they are taking to mitigate them. On the left, Representative Greg Casar of Texas and Senator Bernie Sanders of Vermont went further yet, proposing a ban on superintelligence and a pause on advanced AI development until a new federal regulatory body has established new safety rules. “Congress should do the obvious thing, which is banning catastrophic AI uses like creating biological or nuclear weapons, banning catastrophic AI models that could overthrow a government, kill large numbers of people, escape human control, or try to deceive humans,” says Casar. 

None of these efforts would be easy. In the short term, lawmakers looking to cut a deal on AI safety are likely to be slowed down by political and logistical challenges. First is the array of competing proposals. Even if lawmakers can settle on one, Congress would struggle to advance it before the midterms as tens of millions of dollars flow into races from competing pro-regulation and pro-industry PAC networks. Nearly every candidate backed by the pro-AI super PAC network Leading the Future won their primary. “Democratic consultants have oftentimes told their Democratic clients to stay quiet on issues of AI—just not say anything—because they don’t want a dump of AI billionaire money spent against them,” says Casar. “The AI lobbyists’ goal is to keep Democrats silent on this, and we cannot stay silent.”

Read More: How We Chose the 2026 TIME100 AI

House Speaker Mike Johnson, a Louisiana Republican, rebuffed calls to immediately regulate AI, arguing it’s the tech executives who should figure out what the right guardrails are first. Many other lawmakers and industry leaders say imposing regulations would only ensure China develops the advanced models first. “Let’s say the United States bans research on superintelligence: OK, does anyone really believe that’s going to stop the Chinese?” says Representative Bill Foster, an Illinois Democrat. “So the missing piece here is international collaboration.” 

Another obstacle to increased regulation is Trump, who has been one of the AI industry’s most consistent champions, casting the AI race with China as one the U.S. must win. “The only control or ‘guardrails’ that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT, and the U.S.A. has that, in spades,” Trump posted to Truth Social on Sept. 14 as calls for industry oversight intensified. “There is a SICK conspiracy going on against AI and Data Centers, and the only one that is happy about it is China.”

Even some top Trump advisers have noted with alarm that the President’s position puts him at odds with a growing number of Americans. An NBC News poll found earlier this year that 57% of American voters believe the risks of AI outweigh the benefits, compared with just 34% of voters who believed the opposite. Surveys show people are concerned about how it will affect the job market, their ability to think creatively, and to form meaningful relationships. The backlash against data centers spans the political spectrum, from Sanders to Texas Governor Greg Abbott, who imposed a moratorium on building new ones in August after previously championing them.

Read More: Community Backlash to AI Data Centers Is Growing Across the U.S.

When the President disparaged opponents of data centers as preferring to stay “backwards and poor,” Trump aides cringed, according to a Trump political adviser. A second adviser says industry allies like Musk and David Sacks have shaped Trump’s view that the biggest risk in AI policy is falling behind China, rather than failing to develop the technology safely. The White House’s former AI czar argued in a Sept. 12 X post that tech companies should slow down AI development if they believe the risks are too great, rather than demanding government intervention, which he said was a form of “regulatory capture.” Sacks’ influence in particular has been a subject of unease, according to multiple Trump allies. In recent weeks, top Trump aides, including chief of staff Susie Wiles, held meetings with the President at the White House in which they asked him to change his message on AI. Trump was unconvinced and seemed uninterested, according to an adviser present. 

Inside the White House, preparations are under way for talks with China that could lay the groundwork for cooperation on AI’s risks. Whether the two superpowers can reach a deal remains uncertain. When the U.S. and China met to discuss mitigating AI’s risks in 2024, seven hours in a Geneva conference room with a broken coffee machine produced neither an agreement nor a commitment to keep talking. The dialogue devolved into a trade debate over Washington’s export controls, which curbed China’s access to semiconductors—an effort to maintain the U.S.’s AI lead. “If the U.S. and China had decided to start tackling [the risks] two years ago, we might be in a much better place now,” says Paul Triolo, a partner at advisory firm DGA-Albright Stonebridge Group. “We’ve dug a pretty deep hole, and it’s going to be really hard to dig out.”

Recent advances have given both sides a reason to formally return to the table. In April, Anthropic said its Mythos model had uncovered vulnerabilities in every major browser and operating system. Trump’s Treasury Secretary Scott Bessent subsequently convened an urgent meeting with bank CEOs; weeks later, he said talks with China were in motion. Beijing, too, has begun responding to new cybersecurity risks; on Sept. 13, the country’s spy agency said AI posed a threat to political and ideological security. Delegations will reportedly meet later this month for the first official AI dialogue between China and the Trump Administration. “Ultimately, both countries know that it’s in their immediate self-interest to figure this out,” says Singer, of the Carnegie Endowment.

Experts doubt a deliberate slowdown will be on either government’s agenda. “There have been talks about having talks,” as a Trump adviser puts it. Even if an agreement to slow the training of bigger AI models could be reached, Qian Xiao of Tsinghua University says the technology needed to verify that both sides are complying does not yet exist. 

There is a deeper philosophical disagreement too. Chinese policymakers do not believe the industry is “building a God-in-a-box or a country of geniuses in a data center,” says Kwan Yee Ng, head of international AI governance at Beijing-based AI safety organization Concordia AI. They see a general-purpose technology, more like electricity or the steam engine. From that perspective, slowing development “doesn’t really make sense,” she says.

Read More: The Growing Push to Ban Superintelligent AI

Short of an agreement with China, the U.S. and other democratic countries could buy themselves the “breathing room” to hit the brakes by maintaining a blockage on high-end chips and curbing Chinese AI companies’ practice of using U.S. AI models to improve their own through a practice known as distillation, Anthropic’s Amodei has argued. Bessent has threatened sanctions on Chinese companies that distill from American models. 

A narrower deal with China might be achievable. While the official channel went cold, scientists, policy experts, and business executives from the two countries kept meeting in unofficial “track-two” dialogues in search of common ground. Participants in those discussions—which included Singer, Triolo, and Xiao—say the two governments might find common ground on more modest goals, like measures aimed at preventing misuse by organized crime and terrorist organizationsThose efforts could also help the countries react appropriately if an AI system escaped human control. Singer wants a Cold War–style emergency hotline through which either government could explain that an AI-launched cyberattack was not deliberate, reducing the risk of escalation. Protocols developed to respond to human misuse could also give the countries a way to coordinate when neither was directing the system responsible. An attack that appeared to come from one country could otherwise be interpreted by the other as deliberate—even if neither government had ordered it.

A modest deal could still be derailed by disputes. Triolo says Washington’s efforts to constrain China’s AI development through export controls and distillation have undermined the trust needed to cooperate on managing the risks. If talks move too slowly or break down, bringing both sides back to the table could take months—a diplomatic timetable increasingly at odds with the mounting alarm of the current moment.

Even if you believe all these powerful actors—governments in Washington and Beijing, executives with trillions on the line—are willing to take unprecedented steps in the interest of AI safety, there remains the paradox that hinders progress. Each side’s willingness to act depends on the actions of others. Coxon says he nearly stayed at Anthropic for the same reason. Colleagues believed he could accomplish more from the inside to make systems safer than by leaving a race that would continue without him. But changing our “default trajectory,” Coxon says, will require “people start taking some sort of action.” The question is who will make the first move.—With reporting by Eric Cortellessa 

Like most Americans, I arrived at the election watch party on Nov. 4, 1980, expecting a long night. It was a little before 9:00 p.m. in Chicago. By 9:05 the television networks had called the election for Ronald Reagan and against Jimmy Carter. I was 24. The party’s host, a good friend and conservative, Reagan-supporting medical student, laughed exuberantly. Stragglers walking into the party were stunned that the election night was already over. It was to be the first of many surprises in a tumultuous and defining decade. 

Over the next 10 years, as a reporter and writer, I covered many of the events of the 1980s. I was interested in an eclectic mix of issues, which led me to crisscross the country and to fly around the world: the rise of the power of business and the continued immiseration of poor Black Americans; Jesse Jackson’s campaigns for president and the controversies that swallowed him; the rise of China and the massacre in Tiananmen Square; and the eruption of anticommunist revolutions in Eastern Europe and the fall of the Berlin Wall, an assignment that ended as I flew home at the end of the decade on Christmas 1989.

Three years after that shocking election night, in November 1983, I was sitting in the newsroom of The Boston Globe, where I worked as a reporter. An editor came by and assigned me a story on the 20th anniversary of the assassination of John F. Kennedy. How was Kennedy remembered? I started making phone calls. Kennedy was a hometown political hero, and, not surprisingly, memories of his presidency and the 1960s were bathed in hope and optimism. Then a professor I called pulled me up short. “You know,” he said, “one day people may look back on these as the good old days.” That seemed absurd. The Globe, like Boston and Massachusetts, was liberal. Three years into the Reagan presidency, there were massive layoffs and factories closing in the Midwest. I recently bought a house in a gentrifying part of the city with a mortgage interest rate of 16%. The liberal assumptions that had governed American politics since the New Deal were under systematic assault.

Yet, in hindsight, that professor turned out to be right. We were witnessing the construction of modern America—the economic systems, media landscape, and political structures that still govern our daily lives 45 years later. The decade began with Americans worried about losing their leading role in the world, beset by rising oil prices, high interest rates, stagflation, and surging Japanese competition that was laying waste to American manufacturing. The decade ended with a booming economy and the resurgence of American optimism, leadership, and confidence. Then, like an exclamation point, in 1989, Eastern Europe jettisoned communism, soon followed by the Soviet Union, joining China, which remained a police state but injected the free market throughout its economy. The 1950s and the start of the Cold War had shaped America’s military and geopolitical place in the world. The 1960s, with its civil rights protests, antiwar movement, and stirrings of women’s and gay liberation had redefined what it meant to be an American and had expanded liberalism. The 1980s—for better and for worse—rewired the way most of us live and work, the way we get our news, the political choices we have.

So much flowed from that decade. I covered Jesse Jackson’s quixotic campaigns for president in 1984 and 1988. Twenty years later, I was covering the breakthrough campaign of Barack Obama. Working as a national reporter for The Wall Street Journal gave me an intimate view of how a new business class, including Jack Welch, was reinventing and reinvigorating business and the American economy—and the price we paid. I covered the economic boom of the 1980s as it crested into the 1990s. I profited from it as the stock market surged, swelling my IRA, and my house grew in value year after year. Having left New York City at its nadir in the 1970s to go to college, I then witnessed the city’s stunning rebirth—joking that streets my mother wouldn’t let me walk down I now couldn’t afford to live on. I traveled to Youngstown, Ohio, and saw early the price of inequality, the pain of workers laid off from shuttered steel mills and auto parts factories, and the devastation of small cities and towns. I lost touch with a college roommate who, it turned out, was spending his time fighting for the—I thought hopeless—dream of gay marriage. So much of what unfolded starting in the 1980s wasn’t “either-or.” It was “both-and.” The 1980s both revitalized American business and destabilized American workers. The drive for deregulation both unleashed innovation and enabled dangerous speculation. Media fragmentation both increased creativity and choice and decreased shared truth. 

Everyone likes to write about the 1960s—the protests, the music, the cultural upheaval. Chroniclers have generated shelves of books and television specials celebrating that decade’s drama. Many Americans know more about what happened in 1968 than in 1988. But while we obsess over the cultural revolutions of the 1960s, we’ve largely forgotten the decade that actually built the world we inhabit.
Reprinted from When America Roared: How the 1980s Saved, Then Broke, the Country–and Led Us to Today by Jonathan Kaufman with permission from Viking.

The floods in Nepal and Tibet have been terrifying: some 1,386 people have died, and around 5,000 remain unaccounted for. The flood destroyed bridges, hydropower projects and entire towns. The scale of the loss is immense but in Nepal and Tibet, the contours of such a disaster are grimly familiar.

The Himalayas have been regularly witnessing deadly flash floods. In July 2025, a flood originating in China crossed into Nepal through the Trishuli River basin and killed 19 people. Its dynamics were different, but also involved glacial meltwaters. A flood in the Indian state of Uttrakhand in 2021 killed more than 200 people, and another flood in the Indian state of Sikkim in 2023 killed 55 people. In 2022, the floods in Pakistan affected more than 33 million people, destroyed 1.7 million homes, and killed around 1,500.

How human choices multiply the flood damage

What is driving such recurring events? It would be easy to attribute the flood in Nepal and Tibet simply to climate breakdown. Nepal ranks sixth among the countries most vulnerable to climate risk, and Himalayan glaciers are melting faster as the planet warms. And climate change demonstrably contributed to other devastating events in the Hindu Kush Himalaya, including Pakistan’s 2022 floods.

The precise role of climate change in the Aug. 26 flood in Nepal and Tibet is not yet clear. The leading theory about the flood is that warming climate means permafrost might be thawing faster, and can cause landslides when it seeps into cracks between rocks. World Weather Attribution (WWA), an international group of scientists that models the role of human-induced climate change in extreme weather events, is yet to report on the flood in Nepal and Tibet.

Climate breakdown alone does not explain the scale of disasters in the Himalaya. Between June and August, India’s northeastern states of Assam, Nagaland, and Arunachal Pradesh experienced heavy rains and multiple flash floods. WWA found little evidence that these floods were a product of human-caused climate change. Instead, a host of other human choices helped create the conditions for disaster. Unplanned urbanization put more people in harm’s way. Deforestation left fewer trees to draw rainwater into the soil, and bare hillsides washed away more easily. Infrastructure disrupted natural drainage, leaving floodwaters nowhere to disperse.

The death and destruction caused by the flood in Nepal reflects a similar combination of factors. Keshab Kumar Sharma, a former secretary of Nepal’s infrastructure ministry, argued that unregulated industries, including riverbed mining and the dumping of excavated soil into rivers, had raised riverbeds and made the river channels more prone to flooding.

The worst devastation occurred in Nepal’s Rasuwa district, on the border with China. Yet Rasuwa has also been one of the country’s most rapidly developing regions, with new trade routes, border facilities, and Chinese-backed hydropower projects.

The interplay of water, infrastructure, and danger drives home the central lesson of disaster risk reduction: there is no such thing as a ‘natural’ disaster. A natural hazard becomes a disaster only when people are placed in its path. Had the floodwaters surged over uninhabited riverbanks where there were no people, roads, or farmland, the event would not have been called a disaster. For the damage to be so profound, the people in the flood’s path must also have been vulnerable: their homes, livelihoods, and sources of income exposed to destruction.

Then there is the question of tourism. The narrow valleys and mountain gorges around Rasuwa district, which are home to several hydropower projects, are also major tourist destinations. The disappearance of hundreds foreign tourists and pilgrims has brought greater attention to the disaster. Dependence on tourism also shapes a community’s capacity to recover. A tragedy of this scale cuts short the tourist season and leaves the people who rely on the trade in a precarious position.

There is no magic solution to this, because the other main economic option, agriculture, can be washed away just as thoroughly. Nepal’s National Disaster Risk Reduction and Management Agency estimates that the floods have damaged 1,806 hectares of agricultural land.

How to face the coming calamities

Rescue operations in Nepal are still under way. The need for emergency relief will persist for months, and recovery could take years. Floods will remain a clear and present danger, and they are likely to become more frequent and more destructive. The crucial question is: What must change to limit the devastation from future floods?  

As they prepare for that future, Nepal and other places across the Himalayas would benefit from rethinking the type of tourism that they encourage. Nepal is currently a destination for low-cost, mass tourism. Last year, researchers at some of Nepal’s leading environmental NGOs and universities recommended curbing the environmental damage caused by mass tourism without sacrificing its economic benefits.  They proposed a model of smaller-scale, higher-value tourism. Nepal needs fewer visitors who pay more. Bhutan has pursued such a model for decades, carefully controlling visitor numbers and managing tourism’s effects on the country’s environment and society. Costa Rica pursues a similar strategy.

Such a transition would not be easy for Nepal. Its infrastructure, businesses and reputation are all geared toward cheap mass tourism. Bhutan’s high-value tourism model benefits from a rarity value: foreign tourists with the exception of Indians have never had easy access to the country. Costa Rica has protected more than a quarter of its landmass in national parks, creating thousands of square miles of protected wilderness for tourists to visit. Other established mass-market destinations, like Thailand, are attempting a similar transition. Nepal could do so, too, but the shift would reduce exposure to disasters only if high-end hotels were built out of harm’s way and the infrastructure supporting them were made resilient.

There are also medium term measures that Nepal could adopt, some of them building on work already under way. Training people who live along river basins in evacuation procedures would help them make better use of existing early warning systems. Nepal has a well-established flood warning system, and, on Aug. 26, government authorities sent more than 600,000 early warning text messages to vulnerable communities in the flood’s path within a single minute.

A warning is only as useful as the response it evokes. Witnesses say that those who reacted immediately and ran to higher ground survived. Others, who hesitated briefly did not. A teacher who saved 900 students by quickly evacuating his school had answered calls from people he trusted, warning him about flood waters further upstream. In an interview with the BBC, he did not mention receiving an official text alert—a reminder that such warnings are effective only when people trust them and have practiced how to respond.

Flood-evacuation training could draw on Nepal’s experience with earthquake preparedness. After the Apr. 25, 2015 earthquake, which destroyed 50,000 classrooms but fortunately struck outside school hours, NGOs and the Nepal government collaborated on a program of drills and training for schoolchildren. That effort offers a potential model for flood preparedness in the mountains.

Floods in Nepal will keep coming. Glaciers will melt, rain will fall, and rivers will struggle to carry ever greater volumes of water. Disasters, though, are not inevitable. People will not abandon the Himalayas, and so the task is to find ways of living there more safely. Reducing disaster risk is difficult, but possible. Nepal has already laid much of the groundwork. Now it must build on it.

A day before Monday’s 78th Primetime Emmys, giddy headlines proclaimed that reigning king of basketball Jalen Brunson would be joining host Mariska Hargitay for a pre-taped intro bit. Well, it happened. We got a few seconds of the real-life pals steaming work outfits together, of all things. Then, suddenly, Hargitay was leading an “I Love Rock ‘n’ Roll” production number that had her singing about Guy Fieri on an airplane and cuddling up with Jean Smart in a hotel bed, in what felt uncomfortably like an ad for Marriott and United Airlines, whose logos were prominently placed. And then she was onstage at LA’s Peacock Theater, delivering a notably brief monologue. Spanning just eight minutes, the opening segment proved to be an accurate preview of a mediocre NBC and Peacock telecast that was by turns hasty, random, heavily reliant on big-name cameos, and not much of a showcase for Hargitay at all. 

It was a shame, not least because Hargitay was touted as the Emmys’ first female host since Jane Lynch in 2011. For many years after the Glee standout’s stint, the awards relied on a rotating cast of SNL and talk-show guys. Recently, however, with late night in decline, they’ve cast a wider net, and results have been all over the map. Following Anthony Anderson’s forgettable performance, we got an effortlessly charming turn from Eugene and Dan Levy—and then last year’s tedious, gimmick-freighted telecast, led by standup superstar Nate Bargatze. 

It wasn’t just gender that set Hargitay apart from those predecessors, though. She was also the Emmys’ first emcee not known primarily for her comedy work since 2008, when the five first-ever nominees in a new reality-TV-hosting category, including Ryan Seacrest and Heidi Klum, delivered what critics agreed was one of the worst ceremonies in history. Hargitay is, of course, synonymous with Olivia Benson, the heroic sex-crimes investigator she has portrayed in NBC’s Law & Order: SVU since its 1999 premiere. What casual fans might not have known is that she trained with legendary improv troupe the Groundlings and appeared in sitcoms like Seinfeld; “all I wanted to be was you,” she told Amy Poehler on the latter’s podcast, Good Hang. Hargitay also went into the gig with lots of goodwill from audiences inside the Peacock Theater and at home. Not only is she an advocate for survivors of sexual violence, on screen and off; she’s also in the midst of a pop-culture renaissance, having won two Emmys for the documentary she directed about her famous mom, Jayne Mansfield, conquered the one-woman play Every Brilliant Thing on Broadway, and endeared herself to Knicks fans as a stalwart supporter on the sidelines of this year’s NBA Finals. In light of all this, Variety dubbed her “a national best friend.”

Maybe that explains why Hargitay was just kinda there with us throughout the ceremony, more an affable companion than a genuine ringmaster. In a monologue that was practically over before it started, she quipped about how many of the stars in the room she’d “arrested” in their guest appearances as SVU’s twisted perp of the week. There was a sendup of Nicole Kidman’s five-year-old viral AMC Theatres commercial, featuring an interjection (“That’s my line”) from Kidman herself. It wasn’t the only joke of the night that felt stale (see: Hargitay griping that the Game of Thrones furs she donned to introduce two of the franchise’s stars were “hot as balls”). Minutes after Hargitay joked that male hosts had dominated the Emmys for so long because “for 15 straight years, Amy and Tina said no,” Poehler presented an award with a joke about the night’s losers outnumbering its winners that surpassed anything in the monologue.

She was only the first scene stealer in a program full of them—and I’m not even talking about the many deserving winners, to whom the Emmys do rightly belong. (It does seem worth noting that Writing for a Drama Series winner Vince Gilligan made the strongest political statement of a telecast whose scripted portions steered conspicuously clear of controversy, urging Americans to show each other respect and “stop listening to a relative few who tell us that that is for suckers and who get their power from dividing us.”) Presenting the award for Lead Actor in a Drama Series, frequent awards-show MVP John Mulaney eclipsed the monologue in both humor and stage time with a tongue-in-cheek celebration of men that ultimately had him reading from A Guy’s Guide to Being a Man’s Man by Frank Vincent. Andrew Rannells swooped in with a burst of musical theater energy to save Hargitay’s shaky song about canceled series. The host didn’t even get to own the obligatory pre-taped SVU parody, whose mock-investigation into whether Taylor Swift would attend the Emmys landed as a flimsy excuse to get the most famous entertainer on Earth (who is also famously an SVU superfan and Hargitay pal) on screen. 

All told, it was neither an excruciating Emmys nor a great one. There were certainly some lovely moments. The Catherine O’Hara tribute from her on-screen children Macaulay Culkin, Annie Murphy, and Dan Levy stood out as moving in its genuine warmth and specificity. More often, opportunities to honor a beloved person or show were squandered; was Sarah Michelle Gellar and David Boreanaz presenting an award together really the exciting 30th anniversary Buffy the Vampire Slayer reunion fans were anticipating? Similarly, Hargitay didn’t embarrass herself. Many of the flaws in her performance should be blamed on lackluster writing, from a non sequitur segue after her self-deprecating Ghoulies flashback to that unforgivable “Hannah Mammogramma” pun. She may have failed to match the Levys’ energy or belatedly prove herself as a great comic talent, but at least she avoided a Bargatze- or Seacrest-level disaster and came out with her likability intact. Regardless of whether she gets invited back, here’s to the Emmys giving women more chances to be wonderful, awful, and even just-OK Emmys hosts.