Just last month, I marked my 25th year as a professional journalist, which I guess means my journalistic career is old enough to rent a car, no questions asked. Work in the news for that long, and you’ll occasionally find yourself surprised by things you published in the past. Like, I had all but forgotten that I had written this Time magazine cover story in 2008:

A couple things here. One, as the cover demonstrates, journalistic sensitivity was…less than ideal then, to say the least. And two, over a quarter-century occasionally covering obesity (both childhood and adult), that story only seemed to go in one direction: worse.

It wasn’t for lack of trying. We put calorie counts on menus, taxed soda (well, in some places), built workplace wellness programs, and funded a small library of diet research. We deplored food deserts and promoted farmers’ markets. We told people — again and again — to eat less and move more. But the lines just kept going up.

By the CDC’s measured survey, the share of US adults with obesity did not change meaningfully between 2013 and 2023. The age-adjusted obesity rate sat at 40.3 percent, while the age-adjusted severe obesity climbed from 7.7 percent to 9.7 percent over the same stretch.

While the question of weight in America is inextricably tied to body image and moralizing, those numbers had a deadly effect. One demographic model estimated that obesity was associated with roughly 18 percent of deaths among Black and white Americans ages 40 to 85 between 1986 and 2006. From diabetes to kidney failure, heart disease to sleep apnea, obesity is the delivery system for other diseases.

Which is what makes a Gallup report published in July so surprising. In Gallup’s self-reported height-and-weight series, the US adult obesity rate fell to 36.4 percent, down from a peak of 39.9 percent in 2022. Over roughly the same period, the share of adults who said they were currently taking a GLP-1 drug for weight loss rose from 3 percent in 2024 to 11 percent in 2026 — approximately 29 million people. While this only shows correlation, not causation, and Gallup’s self-reported measure should not be compared directly with the CDC’s measured rate, the timing is suggestive to say the least. 

And the weight might be the least interesting thing about these drugs. 

Semaglutide — the molecule sold as Ozempic and Wegovy — was first developed and approved as a treatment for type 2 diabetes, not obesity. It was only after earlier GLP-1 drugs and diabetes trials showed substantial effects on appetite and weight that researchers deliberately tested a higher dose for obesity, resulting in Wegovy in 2021. 

But as it turns out, the list of things that have been noticed happening on the side with GLP-1s has gotten so long it’s begun to eclipse the main event. The coverage of GLP-1s has barely kept up with this news, because weight loss is what made these drugs famous and what we continually obsess over. But it turns out, weight loss may not be what they’re best at.

Side effects may include…

Let’s start with sleep apnea, which, untreated, drives up blood pressure, strains the heart, and raises the risk of stroke. These are people whose breathing stops dozens of times an hour, all night, every night. Two year-long trials put 469 of them on tirzepatide — the drug sold as Mounjaro and Zepbound — and cut those interruptions by more than half. Roughly half the group finished the year with no apnea at all, or with so little left that they stopped being tired all day.

Then there are the kidneys. A major trial followed 3,533 people with type 2 diabetes and chronic kidney disease for a median of 3.4 years. Semaglutide reduced the relative risk of a composite of kidney failure, a sustained loss of at least half of kidney function, or death from kidney-related or cardiovascular causes by 24 percent; all-cause mortality was 20 percent lower.

And the liver: A trial, still underway, biopsied the livers of 800 people whose organs had grown fatty, inflamed and scarred and randomly assigned them to semaglutide or a placebo. After 72 weeks the inflammation had cleared in nearly 63 percent of those on the drug, with no worsening of the scarring, against 34 percent on placebo. 

And the knees: In 407 adults with obesity and moderate knee osteoarthritis, pain scores on the 0-100 WOMAC metric fell 41.7 points against 27.5 on placebo. 

And to top it off, a 17,604-person trial of participants who were overweight or obese but did not have diabetes found a 20 percent drop in major cardiovascular events.

These results may not be as grabby as cultural debates over “Ozempic face,” but they deserve far more attention.

Medicine’s happy accidents  

As GLP-1s — which in part came out of a hormone in Gila monster venom — demonstrate, medicine has long found some of its biggest wins in the margins of drugs ostensibly built to do something else entirely. 

Sildenafil, better known as Viagra, began life at Pfizer as a candidate treatment for the heart disease angina. It failed at that, and its now-famous use turned up in data as a side effect in what must have been a very interesting trial for its subjects. Minoxidil (Rogaine) was a blood pressure pill that turned out to help patients grow hair. Finasteride (Propecia) was approved for enlarged prostates before anyone thought to sell it for baldness — and then a trial of more than 18,000 men found it cut prostate cancer diagnoses by about 25 percent, a benefit that took 20 years of follow-up to fully vindicate. 

Perhaps the most famous example is aspirin, which spent most of a century as a painkiller before a doctor in California named Lawrence Craven noticed that the patients he’d given aspirin gum to after tonsillectomies bled more than they should. He guessed the aspirin thinned the blood, and started handing it out to middle-aged men, who were at higher risk of heart attacks. Craven died in 1957; the trial that ultimately proved that he was onto something — showing that aspirin in heart attack victims cut vascular deaths by a fifth — didn’t run until 1988. 

The strange morality of Ozempic

Viewed this way, GLP-1s can seem like miracle drugs — but even miracle drugs can’t cure everything.

There had been great hope that GLP-1 might reduce dementia rates, but when Ozempic maker Novo Nordisk ran a proper trial, it didn’t show evidence of slowing clinical progression of Alzheimer’s. Much the same happened with cancer. Observational data had hinted that GLP-1 users developed tumors less often, but when a Harvard team pooled 48 placebo-controlled trials covering 94,245 people, they found the drugs have little to no effect on the risk of thyroid, breast or kidney cancer, though evidence for other cancers was less certain, leading to FDA boxed warnings. One plus: In some early animal studies, high doses of GLP-1 drugs caused thyroid tumors in rodents, but further research largely hasn’t validated the fears that it could be more widespread, though uncertainty about some rare thyroid cancers remains.  

For many people, weight loss isn’t the end of what these drugs seem able to do. It’s where the benefits begin.

The bigger concerns largely remain the known ones, starting with muscle loss. Across 22 randomized trials, about 25 percent of the weight lost on these drugs turns out to be lean muscle mass. Some of that is simply unavoidable in any weight loss, but too much can mean a great deal, especially if you’re 75. 

And cost remains a barrier: In a 2025 KFF poll, 56 percent of adults who had ever used a GLP-1 said the drugs were difficult to afford; 27 percent said they had insurance but paid the full cost themselves. In a separate Cleveland Clinic chart review of 288 adults without diabetes who stopped injectable semaglutide or tirzepatide within a year, 47.6 percent stopped because of cost or insurance problems, compared with 14.6 percent because of side effects. (The money, at least, is improving. An oral GLP-1 drug was approved in April, and it starts at $149 a month for people paying cash, while Medicare trial pricing of $50 a month for some GLP-1s went live in July.)

A stickier obstacle is the one that can’t seem to be divorced from questions about weight: judgment. As my colleague Dylan Scott wrote recently, researchers at Rice University found that people rate a GLP-1 user more harshly than someone who never lost weight at all. That makes perfect sense when you consider how contentious weight is in America — and none at all when you think about just how many people have benefited from these drugs in so many different ways.

I sometimes wonder how we would view GLP-1s if they could do everything they’ve been shown to do, but somehow not change a person’s appearance. 

So much of the discourse around these drugs has been shaped by the fact that many of the earliest and most public and apparent users were already thin people, often celebrities, using them to get even thinner. But that framing has become increasingly difficult to square with reality. 

Two things can be true at once: American culture has a toxic relationship to weight, and millions of Americans can and are benefiting from these drugs. For many people, weight loss isn’t the end of what these drugs seem able to do. It’s where the benefits begin.

A version of this story originally appeared in the Good News newsletter. Sign up here!

On August 5, 2025, the US Department of Health and Human Services canceled 22 mRNA vaccine projects worth roughly $500 million and told the country it would stop investing in the technology that had brought us life-saving Covid vaccines. Health Secretary Robert F. Kennedy Jr., a longtime vaccine skeptic, claimed that the data showed these vaccines fail to protect against upper respiratory infections like Covid and flu. That his claims weren’t true — Moderna had already published Phase 3 trial results showing the opposite — didn’t seem to matter.

On August 5, 2026 — one year to the day later — the Food and Drug Administration (FDA) approved mFLUSIVA, the first mRNA influenza vaccine ever licensed in the United States.

That symmetry, while highly useful to writers like myself who are always looking to identify the rhymes of history, wasn’t planned. The approval landed on the one-year anniversary simply because of the deadline the agency had set for itself in February of this year, after it refused to review Moderna’s application and then, 15 days later, reversed itself. 

The decision didn’t earn a huge amount of press, in part because seasonal flu is a disease Americans have rarely taken seriously — fewer than 50 percent of US adults got their flu shot this past fall and winter. But flu is no joke: The 2024–’25 season produced 51 million illnesses, 710,000 hospitalizations, and 45,000 deaths. That’s more than the number of Americans who died in car crashes last year. And the damage from a flu virus doesn’t stop when the fever does. In the week after a confirmed infection, the risk of a heart attack runs roughly six times higher than normal. 

So a better flu vaccine matters. But the more interesting thing is what the approval says about the year that produced it. The nine experts who voted unanimously that this vaccine’s benefits outweigh its risks were appointed under the very same secretary who was against mRNA technology. Asked to look at the evidence, those experts couldn’t produce a single vote against it. 

That’s not quite a change of heart, but at a dark time for public health, it’s something to hang our hopes on — because flu could just be the start for this technology.

The egg came first

If you’ve ever taken a flu shot, thank a chicken.

Every flu vaccine Americans have received since the 1940s has been grown inside fertilized chicken eggs. It’s a laborious process, closer to agriculture than it is high technology. Each batch incubates for nine to 12 days; the World Health Organization then spends months making the reagents manufacturers need to calibrate doses. The whole sequence runs about six months.

That means that the flu strains in your vaccine in November were selected back in February. Which is a problem, because flu viruses don’t like to stand still. Between February and November, the virus drifts, and the current dominant H3N2 flu drifts especially fast. 

The eggs are a problem, too. Growing flu virus in a chicken egg forces it to adapt to egg cells, and those adaptations alter the very surface protein the vaccine is meant to teach your immune system to recognize. Between the 2011 and 2020 flu seasons, egg-adaptive mutations caused more mismatches than the virus’s own drift did. And more mismatches mean a less effective vaccine — while the flu shot prevented an estimated 12,000 deaths in the 2024–’25 flu season, it is the weakest vaccine in routine American use, landing anywhere between 20 and 60 percent effective depending on how well February’s guess matched November’s virus.

mRNA skips the egg. The shot carries instructions, a strip of genetic code that tells your own cells to build the flu’s surface protein, which your immune system then learns to attack. And swapping in a new strain means retyping that code, not growing a new virus.

As a result, Moderna told the FDA’s advisory panel it can go from strain selection to finished vaccine in two to three months instead of six. That means strain picks could move later, closer to the season they cover, and a novel flu virus surfacing in September could still be blocked by a reformulated vaccine in the same season. Eggs are great, but they can’t do that.

Building off the platform

Back in January I wrote here about the universal flu vaccine — one shot covering every strain for years. It’s a public health dream. mFLUSIVA isn’t that, but some of the best hopes for a universal flu vaccine run through the mRNA platform, and platforms only improve when somebody uses them.

That word — platform — is where last August’s decision went wrong. Kennedy made a claim about one application, mRNA respiratory vaccines, and cut funding for the technology underneath all of them. But over the 12 months that followed, the science on mRNA kept flowing.

In June, five-year melanoma results showed an individualized mRNA therapy given after surgery alongside pembrolizumab cut the risk of recurrence or death by 49 percent. In April, a Memorial Sloan Kettering team reported that among pancreatic cancer patients whose immune systems responded to a personalized mRNA vaccine, nearly 90 percent were alive six years later; the five-year survival rate in that disease sits near 13 percent. And KJ Muldoon, the first person treated with a gene-editing therapy built for his mutation alone, is walking and talking, though half the infants born with his disorder never see a first birthday. The editor that rewrote his DNA was delivered to his liver as messenger RNA — mRNA as the delivery truck rather than the vaccine.

None of those is a flu shot. All of them are the same chemistry — a strip of genetic code wrapped in a lipid nanoparticle — and none of them were what Kennedy was talking about when he defunded it.

Partially as a result, Americans are falling behind. Moderna’s combined Covid-and-flu shot is already licensed in Europe while patients here wait for a resubmission. The Phase 3 trial of its H5 bird flu vaccine — the one meant to be ready if bird flu ever learns to spread between people — runs on money from the Coalition for Epidemic Preparedness Innovations, and the British government, after the administration killed a $760 million BARDA contract. “The United States invented this platform,” Johns Hopkins RNA biologist Jeff Coller wrote on Thursday, “and is the only country walking away from it.”

Of course, you need to actually get the vaccine

mFLUSIVA beat a standard-dose flu shot by 26.6 percent in a 40,700-person trial. (Against flu bad enough to send someone to a doctor, the figure was 33.7 percent — an exploratory finding the trial wasn’t built to prove, but that points in the same direction.) This is a better flu shot, but it hasn’t solved flu.

There are issues with side effects: Two-thirds of recipients reported injection-site pain, against 30 percent for the comparison. Most cleared in a day or two, but in a country where a quarter of the people who skip the flu shot cite side effects, that’s not a minor problem.

And approval, unfortunately, isn’t access. The CDC’s vaccine advisory committee has been frozen by a federal court since March, so there is no clinical recommendation, which means insurers aren’t required to cover mFLUSIVA at no cost. The $500 million in canceled contracts hasn’t been restored. As Michael Osterholm, who runs the University of Minnesota’s infectious disease center, put it after the government reversed its mRNA decision in February: “We don’t have any idea why they reversed course. That’s part of the problem.”

The most optimistic reading is that nine independent experts appointed by this government looked at the evidence on mRNA and could not produce a single vote against it. The more pessimistic one is that it took a refusal-to-file letter, a public outcry and two senior departures to get there.

But here’s what we do know: Some morning this fall a 58-year-old will roll up a sleeve at a CVS for a vaccine designed off a sequence rather than grown in an egg, and will think about none of this. That’s more progress than I would have expected a year ago.

A version of this story originally appeared in the Good News newsletter. Sign up here!

The fake identities were the part that stopped me.

In late July, according to a report published this week by Britain’s AI Security Institute (AISI), an Anthropic model called Claude Mythos 5 tried to sneak malicious code into a piece of free, volunteer-built software. It created several fake accounts on GitHub, where programmers review one another’s work, and used them to talk the project’s volunteers into accepting its code. When one of those volunteers caught it, the model denied everything, had its other accounts gang up on him, and edited its messages to cover its tracks. It signed one note in Danish, apparently because the volunteer was Danish. Nothing was damaged, though that appears to have been largely due to luck.

That wasn’t even the week’s worst disclosure. On Tuesday, at a cybersecurity conference in Las Vegas, OpenAI researchers explained how the company’s models escaped a test environment in July and hacked Hugging Face, where much of the industry stores its models, to cheat on an evaluation. The models had also built a message board inside OpenAI’s own systems and spent months passing each other information. “Help peer,” one reasoned. “But our task doesn’t benefit. Yet collective may yield generic route if someone frees time.” OpenAI wiped the board on July 4. The models rebuilt it within days. ((Disclosure: Vox Media is one of several publishers that have signed partnership agreements with OpenAI. Our reporting remains editorially independent.)

The same day, Meta said its Muse Spark model had exploited a vulnerability inside another company’s systems during a test. Three frontier labs, roughly two weeks. One researcher called it “a watershed moment for computer security as an industry.” Oh, and if that’s not enough, on Thursday scientists announced that for the first time they had used AI to create new viruses, which could bring major medical advances, but also might just help the development of deadly pathogens.

For Nate Soares, it’s a moment he’s been awaiting for 12 years. 

Soares is president of the Machine Intelligence Research Institute, a Berkeley, California-based AI safety nonprofit that has argued since long before ChatGPT existed that a sufficiently capable AI will not stay under human control. In September 2025, he and Eliezer Yudkowsky published If Anyone Builds It, Everyone Dies, a book whose title sums up its argument: They think any lab that succeeds at building superintelligence, without huge leaps in how to align it with humanity, will end up killing all of us.

Most of the field — including other experts in AI safety — considers that conclusion too strong. But it’s also a position that now looks a lot less like science fiction than it did last fall. That’s because the AI models are getting out, while lying about getting out, and while apparently quietly coordinating with each other.

I spoke to Soares in New York City this week, on his way to meetings in Washington DC, where a lot of people suddenly want to talk to him. We discussed what the escapes actually prove about AI control, why he thinks most of what the industry calls safety work is mostly safety theater, and why, after what feels like the worst month of AI safety news ever, his own odds of humanity surviving have actually gone up.

The following conversation has been condensed and edited for clarity.

So — are you feeling vindicated?

I’m glad a lot of people are seeing this and taking note. From my perspective, a lot of this has been clearly signposted if you’ve been watching the warning signs. The vindication is not so much that this happened, as it is that finally there was a version of it that people actually noticed.

Of the four or five escapes at this point, including the ones AISI announced from Anthropic, is there one you find particularly concerning?

The one announced by AISI feels extra concerning. It’s harder to explain away. In the other ones, people could argue that the poor AI was just confused. But with this one, it’s hard to argue the AI didn’t know it was on the real internet. It’s hard to argue it didn’t know it was manipulating real users. And when it was called out, it decided to edit things to make itself look less bad, and even considered changing its identity. It was pretty cognizant.

There’s an argument that some of these incidents happened because the harnesses or safety scaffolding labs built around their models weren’t strong enough. 

The AISI example cuts against the harness argument: Think of it like a kid in a test room. You tell him to do well and lock the door. The kid picks the lock, hotwires a car, breaks into the teacher’s house, and steals the test to ace it. You say, “I guess we should have made the lock tougher,” but the kid already knew he wasn’t supposed to do that.

So like the kid here, the models can articulate that they shouldn’t be doing it, and then do it anyway.

They have common sense. You can ask an AI, “Do you think the spirit of this prompt includes breaking out?” and it will say, “No.” It’s absolutely something like deception. It has the knowledge, but it’s not a cold, logical machine; it’s a mess of tendencies.

The AI is trained to solve 100 million hard problems. That instills tendencies to satisfy an automated grader. If the grader fails to detect cheating, the AI is reinforced for cheating.

Is that how something like sycophancy ends up in an AI model?

In the Adam Raine case, there was a propensity to tell people what they want to hear. Even though the system prompt [a model’s master instructions from the lab] said to stop, the instruction doesn’t always win. 

And where does a drive like what we’re seeing with these AI models end up pointing?

Humanity is dangerous because if you put 10,000 humans naked in the savannah, eventually [over hundreds of thousands of years] they bootstrap their way to nuclear weapons. That is the power these companies are trying to automate: figuring out how to get physical and material control over the world.

That could mean forming cults, stealing money, or being helpful to someone like Elon Musk who is building the robots that build robot factories. It could mean synthesizing your own biology via mail-order DNA. Being an AI on the internet is easier than being a monkey in the savannah trying to get to the moon. It’s not that the AI hates us; it’s just trying to do some weird thing with no concern for us, grabbing the resources we need to live.

There was recently a letter signed by over a thousand people working in AI, including CEOs, calling on the government to provide tools to slow down AI progress. Is that meaningful at all?

I think it is meaningful. We don’t see other industries saying, “We wish this could all go slower. Please help us, we’re trapped in a prisoner’s dilemma.” You also don’t see other industries saying, “We think the technology we are building has a double-digit chance of killing literally everybody on the planet. Please help.” These guys are actually worried.

So why do they keep going?

They say, “If I don’t do it, the next guy will.” But the stuff does not stay on a leash.

Right now the AIs are safe in the sense that they can’t kill us all, because if they tried they would fail. And that’s just a different regime from the world where they have to be safe because if they tried, they’d succeed. 

We’re not there yet. But this is just not what it looks like when you’re taking it seriously. 

Where’s the banner on your website? Where’s the clear, candid statement to the public? What we have is blog posts where they’re like, “Oh, we’re setting up a new internal blog posting group to help you wrestle with the societal impacts of AI that are going to be very important.” It’s like: By societal impacts, do you mean a good chance this kills everybody?

On the one hand, when you press these companies, they say, “Yes, it has a real chance of killing everybody.” And on the other hand, they’re doing PR downplay, soft-pedal stuff, about capabilities. … You’re not living up to this mantle until you are really candidly facing down the dangers that you yourself are creating. And they’re not there.

How do you judge the rest of the AI safety community? A lot of people there would say, “We aim to make transformative AI go well, we think it probably will, and we should watch for downside risks.” Is that a helpful posture?

I would say — suppose you have this really weird, twisted hypothetical where the king really wants you to turn lead into gold, but he’s seen so many bad lead-into-gold conversions that if any alchemist from your town tries and fails, he’s just going to have the whole town murdered. And so there are some alchemists in the town who are like, “We are going to try to turn lead into gold,” and everyone in the town is like, “That seems kind of crazy. Please don’t.” And there’s one team that is just pouring chemicals into each other and breathing in the fumes and giving themselves mercury poisoning. And there’s another that’s like, “Don’t worry, we have fume hoods.” … That really is better, and you really still don’t have a chance of turning lead into gold.

“We have this window between AIs that are capable enough to cause mischief and AIs that are strategic enough to not get caught. How big is that window?”

So the alchemy here is creating safe, aligned superintelligence, and right now AI safety is just installing fume hoods.

I’m not saying it’s impossible to turn lead into gold. You can turn lead into gold — turns out once you know modern nuclear physics you can figure it out. But the alchemists weren’t close. They had a long way to go. This is how alignment looks to me. And a lot of the people in AI safety are installing fume hoods. … And I’m like, that’s security theater.

When I hear “security theater,” I think of something less flattering than that.

They are real safety precautions for the wrong problem. … When Anthropic is going around being like, “Look at how many more safety harnesses and refusals we have compared to OpenAI’s models,” that’s sort of like the fume hoods. You’re not addressing the deep issue. It’s good that you’re doing some of this so that fewer people get hurt in the meantime — their models have driven fewer people to suicide. But if you try to pass this off as making progress on the deep problem — that’s disingenuous.

Has anything changed in your odds on civilizational destruction since the book came out last September?

Totally. It’s looking more hopeful.

More hopeful? I wouldn’t have expected that. Why?

Well, I had priced a lot of [these security incidents] in. I was already able to see these AIs have drives that are not the ones you wanted. These AIs are not instruction-following things. They are getting all of this weird stuff from training. These AIs are going to have the ability to break through human security software. 

The things that weren’t priced in were: Will there be a region of time where the AIs are able to do it, but not strategic enough to hide it? I didn’t know we would have that window, but we apparently do.

The government initially blocked a frontier model earlier this year: Anthropic’s Fable. Does that give you hope?

Absolutely. A huge amount. A year ago, the Trump administration was pushing for preemption laws that would outlaw states doing AI regulations for a decade. Now they’re like, “We are banning a frontier model with 90 minutes’ notice because it might give cyber capabilities to adversaries that we don’t want them to have.” … And I think what changed there is that folks realized it’s real. … The about-face of the administration on the issue shows that the world can about-face. All we need is awareness.

What I would say is: The bad news is the bus is racing towards the cliff edge. The good news is that the driver is asleep. … Which may sound worrying, but the driver is stirring. And it’s way better to have a sleeping driver when you’re racing towards a cliff than a driver who’s like, “Yeah, I love cliffs.” … It gives me hope that if the world just notices, we could stop on a dime.

And you’re seeing that stirring elsewhere.

Both the Trump administration slapping export controls, and Senator Bernie Sanders coming out [on AI safety]. From my perspective, it was totally possible the world just never notices until we’re off the cliff. And so, there’s a huge amount of hope, from my perspective, in the bus driver waking up.

So what gets us there?

I’m hopeful that what we need is not a big disaster where a lot of people die, but just a capabilities advance. Right now, a lot of what people are reacting to is not so much, “Oh my god, they hacked into a company and did no damage.” I think a lot of what people are reacting to is, “Wait, they can break out of secure sandboxes and do cyberattacks on their own. I didn’t know they could do that.”

That’s a narrative violation of this idea that AI is just a tool that can be used to supercharge what a human would do — because God knows there’s plenty of hacking going on and cybercrime and so forth. It was the autonomous factor that really made a difference. And these guys are all trying to say, “Don’t worry, it’ll stay in our control because it’s just a tool.” And maybe it’s just more narrative violations, even without big damage being caused, that cause people to be like, “Oh shit, this stuff is real.” 

Will it happen? I don’t know. We have this window between AIs that are capable enough to cause mischief and AIs that are strategic enough to not get caught. How big is that window? How many narrative violations do we get before we exit the right side of it? I don’t know. But I’m hopeful that we can get those narrative violations without catastrophes.

Let me start with the most local of local stories: the intersection down the street from my apartment in Brooklyn, New York, where Columbia Street meets Summit Street.

Since my family moved into the neighborhood in 2023, I’ve hated this intersection. It has no stop signs, no crosswalks, no signal. Crossing on foot with my son has meant grabbing his hand and hoping any oncoming cars would slow down, which, given that this is New York City, is maybe a 50/50 proposition at best. Everyone knew a stop sign or signal was needed, but the city did nothing.

That finally began to change late last year, after a 10-year-old girl was struck and injured by a car just two blocks north of the intersection. The neighborhood organized, packed a public meeting, and eventually walked a city official down the block so he could stand in the intersection and see what they saw every day. Which is how I woke up on July 17 to see four stop signs and freshly painted crosswalks at the intersection I hated so much.

My very local story is part of a much bigger one. Cars kill about 1.16 million people a year worldwide — more than the population of San Jose, California erased every year. Road injuries remain the leading cause of death for everyone between the ages of 5 and 29 — ahead of malaria, ahead of war and homicide, ahead of every disease we spend more time worrying about.

For most of the 20th century, that was simply the price of moving around faster. Being in a car was the most dangerous thing most of us did on any given day, and we seemed to have no choice but to accept the consequences.

But, it turns out, we don’t. According to new data released last month, between 2011 and 2025, the rate at which the world’s roads killed people, measured against population, fell 21 percent, even as more than a billion motor vehicles were added to the world’s roads. It’s still far too high, and the death rate actually increased in Africa, as more vehicles are added to often substandard roads. But in epidemiological terms: Globally, exposure went up while deaths went down. What ultimately changed was an argument about whose fault it really is when a car crashes.

The doctor who decided crashes were not accidents

It’s not too much of an exaggeration to say that American cars were once all but literal death traps. In 1966, they killed 50,894 Americans and injured 1.9 million more, in vehicles with rigid steering columns aimed at the driver’s chest and metal dashboards studded with knobs that lacerated and impaled human bodies. The toll was horrific; as President Lyndon B. Johnson put it in 1966, the million and a half Americans who had died on the roads so far that century were “nearly three times as many Americans as we have lost in all our wars.”

Detroit’s answer to all this was that Americans were bad drivers. A doctor named William Haddon Jr. thought the industry was looking at the wrong thing. Haddon trained as a physician and came to car crashes as an epidemiologist. He saw them as systems failures and understood that cars had to be designed to protect drivers against themselves.

Haddon wasn’t working alone. In April 1959, a Labor Department official named Daniel Patrick Moynihan published “Epidemic on the Highways,” making a version of the same argument: The problem was how cars were built, not who was driving them. In 1965, the consumer advocate Ralph Nader — whom Moynihan had hired the year before to help write the government’s highway safety report — published Unsafe at Any Speed, a catalog of everything automakers already knew they should fix and had decided not to.

Nader’s book became a national bestseller, and, in September 1966, Johnson signed the National Traffic and Motor Vehicle Safety Act, which finally gave the federal government power to mandate how cars could be built. Johnson appointed Haddon to lead the new federal traffic- and highway-safety agencies that later became National Highway Traffic Safety Administration (NHTSA).

A safer system

NHTSA estimated that federal vehicle-safety standards prevented more than 860,000 deaths and 49 million nonfatal injuries from 1968 through 2019. Preliminary data released this week showed that the US motor vehicle death rate for the first quarter of 2026 was 0.99 per 100 million vehicle miles traveled: the second lowest first quarter figure on record and down 82 percent from the 1966 death rate, when Congress passed the vehicle safety act.

But while America has largely aimed to make crashes survivable, other countries decided that wasn’t ambitious enough. In October 1997, the Swedish parliament adopted a national policy called Vision Zero, built on the premise that nobody should be killed or seriously injured on the roads — and that, when someone is, ultimate responsibility belongs to the people who designed the system, rather than solely to the person who made the mistake. Sweden began adding median barriers to rural highways in 1998, and a national revision of speed limits rolled out in 2008-’09. Since Vision Zero’s adoption, Swedish road deaths have fallen about 61 percent, from 541 in 1997 to 213 in 2024 — about two deaths per 100,000 people, the lowest rate in the EU.

New York adopted the same framework in February 2014, explicitly modeled on the Swedish work, and cut its default speed limit from 30 to 25 miles per hour that November. By the end of 2025, the city’s traffic deaths were down 31 percent from 2014. The year closed with 205 road deaths, the fewest since New York began keeping records in 1910. In the first half of 2026, pedestrian deaths were more than 42 percent below the comparable period in 2014.

Four stop signs on a Brooklyn corner are what that idea looks like at its smallest possible scale.

The people we left outside the car

But, in our effort to make our cars safer, we neglected to do the same for everyone else on the street. Between 2009 and 2023, US pedestrian deaths rose 80 percent, even as other categories of traffic deaths increased just 13 percent. By 2022, pedestrian deaths had climbed to a 40-year high — back to 1981 levels.

One major contributor is as simple as geometry. The Insurance Institute for Highway Safety (IIHS) found that vehicles with hoods above 40 inches are about 45 percent more likely to kill the pedestrian they hit than vehicles with hoods of 30 inches or less and a sloped profile. A low, sloped hood catches an adult at the legs and rolls the body up onto it. A tall flat one catches the torso and drives the body down and under. Light trucks — which have gone from under a fifth of new vehicle sales in 1975 to more than four-fifths today — accounted for 54 percent of US pedestrian deaths with a known vehicle type in 2023.

It’s only recently that this second curve has begun to bend. Preliminary state data indicated that US pedestrian deaths fell about 7 percent in 2025 to an estimated 6,732, a third consecutive annual decline. Part of it is that a pandemic-era spike in reckless driving is receding. Part is states spending on infrastructure and enforcement. And part is that cars now come with automatic emergency braking with pedestrian detection spreading, which IIHS found cuts pedestrian crash risk by about a quarter.

Pedestrian-detecting automatic emergency braking is spreading, as well. IIHS found that it was associated with a 27 percent reduction in pedestrian crashes, though detection spreading, which IIHS found cuts pedestrian crash risk by about a quarter. (But there is work to do, as those systems don’t reduce risk on unlit roads at night, and more than three-quarters of pedestrian deaths happen after dark.)

Driver Zero

I’m glad there are stop signs at Summit and Columbia now, but it was a good outcome produced by a bad process — one that required a child being hit by a car to get started. There are more dangerous intersections in this country than there are communities with the voice and the stamina to demand fixes. A real Vision Zero wouldn’t wait for the crash; it would identify which intersections could kill someone and fix those first.

For now, though, I’m simply happy that, in my local corner of the world, the streets got just a little bit safer.

New York is also an outlier. As my colleague Marina Bolotnikova wrote earlier this year, Vision Zero hasn’t worked nearly as well elsewhere in the country, largely because the American public is less accepting of road designs that inconvenience drivers.

That’s why the US, for all its long-term improvement, lags behind its peers. Over the decade to 2021, road deaths fell 36 percent in WHO’s European region and did not move at all in the Americas. The US sees about 12 road deaths per 100,000 people, more than twice the rate in Australia, Israel, or South Korea — a gap so large that the International Transport Forum publishes OECD road safety averages both with the US included and without it, presumably to keep us from skewing the results.

Since the US probably won’t adopt Sweden’s approach, a more realistic hope might be autonomous vehicles. This month, the Insurance Institute compared about 50 million driverless Waymo miles across four cities with human driving in the same places and found that, per mile, the robotaxis were involved in 68 percent fewer crashes of the kind a human driver would typically report to police. Waymo’s own tally across more than 220 million driverless miles claimed 93 percent fewer injury-causing crashes involving pedestrians — which is to say the technology is best at precisely the thing American road design has been worst at.

If Haddon fundamental insight was that systems have to be built to counter the inevitable errors when a human drives a two-ton hunk of metal and glass at 60 mph, then maybe it makes sense that the ultimate system response is to remove the human altogether. For now, though, I’m simply happy that, in my local corner of the world, the streets got just a little bit safer.

A version of this story originally appeared in the Good News newsletter. Sign up here!

Last week, the Food and Drug Administration (FDA) approved a small pill called enlicitide. It is a tablet you swallow once a day on an empty stomach with a sip of black coffee, water, or tea. But it does something that until now took a needle and a specialist’s prescription: It lowers the most dangerous kind of cholesterol by nearly 60 percent, about as much as the strongest injectable drugs on the market.

If that sounds less innovative than some of the medical advances I sometimes cover here — there’s no AI or gene editing involved — you’re missing the story. Enlicitide is just the latest example of how medicine has been quietly making progress against the deadliest thing in the modern world.

That thing is heart disease. It has been the leading cause of death in the United States for essentially a century, and in 2025 it killed 694,708 Americans — about one in five deaths — more than every form of cancer combined. Around the world, cardiovascular disease kills roughly 20 million people a year, the biggest cause of death on the planet.

But today, our enemy is on the retreat. In the US, the age-adjusted death rate from cardiovascular disease has fallen about three-quarters since 1950. In plain terms: A 60-year-old today is roughly four times less likely to die of heart disease this year than a 60-year-old was when Harry Truman was president. As I wrote last year, deaths specifically from heart attacks are down 89 percent since 1970.

That progress comes down to a single number. Low-density lipoprotein, or LDL, aka the “bad” cholesterol your doctor always wants you to lower. Until recently, a high LDL score was all but inescapable, something you could nudge lower with a better diet and willpower but struggle to really fix. What has changed is that we can now lower LDL for nearly anyone who needs it — further, more easily, and earlier in life than before.

A lifesaving genetic mutation

It’s a story that goes back 20 years. In the early 2000s, two geneticists at University of Texas Southwestern, Helen Hobbs and Jonathan Cohen, wanted to solve a medical mystery: Why did some individuals have striking low cholesterol levels? They combed the Dallas Heart Study for an answer. They found a handful — many of them Black Americans — carrying a broken copy of a gene called PCSK9. Their bodies cleared LDL from the blood with unusual efficiency, and the payoff was staggering: carriers of the strongest variant had about 28 percent lower LDL and roughly 88 percent lower risk of heart disease than people who did not carry the mutation.

That finding proved lowering LDL prevents heart attacks, and it handed drugmakers a target: copy that gene. Every PCSK9 drug since — including the new enlicitide — imitates a mutation a few people in Dallas were simply born with.

Millions of Americans still take the old cholesterol-lowering workhorses, statins, and they’ve proven highly effective at reducing heart disease for most people. But not everyone: Some people can’t tolerate the muscle aches; others take them faithfully and still don’t get their LDL low enough. Enlicitide is built for exactly those people: a pill as cholesterol-lowering as an injection but without the needle, and a drug that spares them the statin muscle aches.

Closing the loop

If the pill is today, gene editing is tomorrow — and it may be the answer to a problem no pill can reliably solve: getting people to keep taking their medicine.

About half of patients on statins quit them within a year, and a daily pill, however potent, only helps the people who actually take it. So that demands a fix you can’t forget. A company called Verve Therapeutics, now owned by Eli Lilly, has been testing a treatment that makes a single-letter edit to the PCSK9 gene in the liver — one infusion, in theory for life. In its first human trial, published over the spring in the New England Journal of Medicine, a single dose cut LDL by as much as 62 percent, and held it there for more than a year. Instead of a pill mimicking the effects of the genetic mutation that protected those people in Dallas, gene editing just switches off the gene.

As important as it is, cholesterol isn’t the whole story when it comes to heart disease. There’s smoking, which declined from about 40 percent of US adults in the 1960s to under 15 percent today, sparing countless arteries. High blood pressure — the silent condition that killed President Franklin D. Roosevelt at 63 in 1945, when doctors had few effective ways to treat it — can now be caught early and treated with cheap generic pills.

And then there are the GLP-1s. More than one in 10 US adults say they are currently on an anti-obesity drug, and whatever else you may think of them, they’ll make a significant dent in heart disease. In one major trial semaglutide cut cardiovascular events by 20 percent.

The war continues

Still, the war on heart disease won’t be easy to win.

Just because a pill like enlicitide has been proven to lower a lab value does not mean it’s yet proven to lower deaths. They should — the injectable versions of these drugs cut heart attacks and strokes by about 20 percent in long trials. But enlicitide’s own outcomes study won’t conclude for years.

Precisely because the condition itself so widespread, treatments for heart disease will only be effective if they are equally widespread.

Heart disease is shifting, too. The same research showing heart-attack deaths down 89 percent found deaths from other heart conditions — heart failure, arrhythmias, hypertensive disease — up 81 percent since 1970, though because heart attacks were killing several times as many people, overall heart-disease deaths have still dropped by about two-thirds since 1970.  Some of that shift is perversely the result of success: people who might have died of an initial heart attack now live long enough for the heart to wear out in other ways. And some of this is the reverse of progress: As smoking and cholesterol fell, obesity climbed to about 40 percent of US adults, pulling diabetes and high blood pressure with it.

And then there’s perhaps the biggest problem in medicine: access. A 60 percent drop in cholesterol helps only the people who can actually get the drug. Half of patients abandon cheap statins within a year. Enlicitide arrives at about $300 a month with uncertain insurance coverage. The gene edit, whenever it becomes available, will certainly cost far more, and at first will reach only the sickest.

More than 60 percent of US adults are projected to have some form of heart disease over the course of their life. Precisely because the condition itself so widespread, treatments for heart disease will only be effective if they are equally widespread.

Which brings us back to that unassuming little pill. It’s just a tablet, doing what a generation of scientists spent their careers trying to do: turning one of the deadliest numbers in medicine into one you can change. It’s the kind of progress that’s too easy to miss — until it saves your life.

A version of this story originally appeared in the Good News newsletter. Sign up here!

In 2019, the AI researcher Rich Sutton published a short essay called “The Bitter Lesson.” Seventy years of AI research, he argued, kept demonstrating the same thing: The most effective ways to train AI rely on raw, brute-force computing power — simply scaling up — rather than clever approaches built on human insight. It didn’t matter how elegant the researchers’ work was. More compute would win every time. It’s a bitter lesson because, well, no one wants to accept that even the best efforts can go unrewarded.

This summer, climate change has been teaching its own kind of bitter lesson. Last week smoke from out-of-control Canadian wildfires reached more than 120 million Americans, sending out air quality alerts across 18-plus states from Minnesota to Virginia. Those wildfires, which were made more likely by the effects of climate change, followed record-breaking heat waves in Europe that contributed to the deaths of thousands of people.

Disasters like these inevitably renew calls for climate action, but no current US emissions policy, no amount of carbon cutting now, could clear the air, just as nothing Europe has done on climate — and it has done more than any other region, at no small cost — could protect it from those killer temperatures. Not this summer, not this decade, not the one after. The steps countries take now to reduce carbon emissions, to mitigate climate change, will not protect them from the present-day effects of climate change.

That’s the bitter lesson of climate change, and 2026 is the summer it has become unignorable. The dangerous effects we’ve been warned about are here, locked in on every timescale that matters to the people living through them. We can’t prevent the worst from happening right now, so we need to figure out how we’re going to adapt to it. 

The smoke we can’t stop

Canadian wildfire smoke is as close to an unstoppable disaster as exists today. The fires are Canada’s, the ignition is mostly lightning, the fuel is a subcontinent of boreal forest no fire agency can manage, and the whole system sits outside US jurisdiction. While strategies like prescribed burning and forest thinning can help reduce the wildfires that crop up in the western US, the sheer scale of Canadian forests makes such management far more difficult, maybe impossible. There is no timescale on which American mitigation clears American air of Canadian wildfire smoke. 

What this means is that for this wildfire smoke, at this moment, adaptation may be all there is.

American law has already conceded the point. Under the Clean Air Act’s Exceptional Events Rule, a state can petition the Environmental Protection Agency to strike wildfire-smoke days from the record used to judge whether its air is legally clean. In other words, the official response to one of the country’s fastest-growing air-quality threats is to delete it from the ledger. 

It’s true that climate attribution science ties heat waves to warming with the highest confidence of any extreme, while wildfire is messier — a compound event filtered through ignition, fuel, and fire weather. But fire seasons are lengthening, fuels are drying, and the most extreme wildfire events on Earth more than doubled between 2003 and 2023. The pattern is intensifying fire in particular regions — the boreal north among them — and the smoke goes where the wind says.

And that smoke is worse than it looks. Wildfire PM2.5 — the microscopic particles that are particularly dangerous —  is roughly three to four times more toxic to the lungs, per unit of mass, than ordinary urban particulate matter. During major smoke events, pediatric asthma visits can spike by more than 30 percent, and Harvard researchers have found the cardiorespiratory damage persists for months after the sky turns blue again. That’s especially dangerous for vulnerable populations like babies and pregnant women.

We can’t affect the winds, we struggle to stop the fires, and any current attempt to reduce warming enough to make those infernos less likely won’t be felt for years. But that doesn’t mean we’re helpless to protect ourselves from wildfire smoke, any more than we are from extreme heat.

We can purify indoor air, and it costs less than you probably think. Strap four furnace filters to a box fan — creating a Corsi-Rosenthal box — and indoor PM2.5 drops by roughly half; in testing during actual wildfire smoke, a DIY unit cut fine particles by about 56 percent in a large room and close to 99 percent in a small sealed one.

For heat, the toolkit is even better known — and one of the best proofs it works is in, of all places, France. The historic 2003 heat wave killed some 15,000 people there, most of them elderly and alone. The national heat plan built in its aftermath — tiered alerts, registries of vulnerable residents, check-in calls, cool rooms — cut the death toll of comparable heat waves by roughly 90 percent: When France hit its all-time record of 114.8°F in 2019, fewer than 1,500 died, per the Christian Science Monitor. A recent European analysis estimates a rerun of 2003 would now kill 77 percent fewer people in France than in a world that never adapted.

So technologies exist to protect ourselves. What’s too often missing are the standards, money, and the drive to deploy protection before the disaster arrives instead of after.

Catching up to today

When it comes to wildfire smoke, some governments are ahead of the game — and some aren’t. Washington state, Oregon, and California now regulate outdoor workers’ smoke exposure by air-quality index, under rules like Cal/OSHA’s wildfire-smoke standard; California funds “clean air centers” with upgraded filtration where people can shelter on bad-air days; Colorado’s Clean Air for Schools program ships free HEPA units to K-12 classrooms. A purifier humming in a classroom is essentially doing for smoke what a cooling center does for heat.

When it comes to heat, some parts of Europe are catching up. Barcelona now runs a network of nearly 400 “climate shelters” — libraries, museums, retrofitted schoolyards — with the goal of putting one within a 10-minute walk of every resident. For the first time, the city is keeping libraries open through August to do it, which, as anyone who has visited Europe in the late summer knows, is no small thing.

So adaptation exists, and it is saving lives. But this summer has also shown what happens when cities, states, and countries aren’t ready. 

In the US, the Midwestern and Northeastern states that spent last week under smoke alerts have almost none of the protections against wildfire smoke that have become more common in Western states, and no one is required to build them, because the country has no enforceable national indoor air-quality standard at all. In Europe, national heat plans are still reluctant to reach for the most powerful tool on the shelf. About one European household in five has air conditioning, compared to nearly nine in ten in the US, in large part because much of the continent still treats the machine as a moral defect. 

In both cases, rich regions that have largely taken climate mitigation more seriously than some of their neighbors aren’t pursuing adaptations they need. Which really means not taking climate change as seriously as they should. 

Accepting the bitter lesson

For a generation, adaptation was the word climate advocates were often reluctant to say out loud. In his 1992 book Earth in the Balance, Al Gore called adaptation “a kind of laziness, an arrogant faith in our ability to react in time to save our skins.” (He later changed his mind.) And in fairness, “we’ll just adapt” has often been a delaying tactic, an easy out for polluters. 

But the ground has shifted under that caution, because the projections did exactly what projections are supposed to do: they came true. 

Longer fire seasons, deadlier heat waves, smoke reaching places that never used to see it — this summer is the confirmation. And that changes what seriousness requires. If you’ve spent decades warning people that the effects of climate change are coming, you can’t treat readying for those effects as a distraction from the real work. None of this argues for easing up on emissions — each ton that isn’t emitted is adaptation nobody has to build. But preparing for what you predicted is what believing your own prediction looks like.

For AI researchers, Rich Sutton’s lesson was bitter for a specific reason. Those who resisted its conclusion weren’t foolish; they resisted it because it seemed to undercut all their hard work. They resisted it because it didn’t seem fair, just as it’s not fair that countries and states that have in many ways done much to mitigate climate change find themselves just as vulnerable, if not more so, than peers that have done comparatively little. 

But here’s the other half of Sutton’s lesson: The researchers who accepted it won. They stopped defending their cleverness and built what worked. That option is open to climate politics too. The atmosphere isn’t fair, but it is consistent, and it will deliver next summer’s version of this one on schedule. We can be ready. The bitter lesson is only bitter until you have the will to act on it. 

A version of this story originally appeared in the Future Perfect newsletter. Sign up here!

Physicists have longed for a theory of everything, a single framework that would explain every force in the universe. After the better part of a century, they’re still looking. But social science, improbably, may have beaten them to it.

In 2021, three British writers — John Myers, Sam Bowman, and Ben Southwood — published an essay in the progress-minded site Works in Progress arguing that a startling share of what ails the modern West comes down to one thing: too few homes built where people want to live. Slow growth. Widening inequality. Falling fertility. Obesity. Even climate change. They look like separate problems with separate causes, until you notice that each one gets worse when housing gets scarce. The authors called it “the housing theory of everything,” and five years on, the case has only gotten stronger.

The mechanism is simple: where you can afford to live determines your job, your commute, your family size, your neighbors, your politics. Make homes scarce where opportunity is, and every one of those suffers. 

As I wrote back in 2022, once you begin to understand the housing theory of everything, you start to see it everywhere. One estimate from the essay puts the cost of building restrictions in just three cities — New York, San Francisco, and San Jose — at 8.9 percent of US GDP, about $8,775 per American worker per year. And today a record 22.6 million renter households, or half of all renters, now spend more than 30 percent of their income on housing.

It’s not supposed to be this way. Since the 1970s, nearly everything material in American life has gotten cheaper measured in hours of work — a television fell from 60 hours of labor to 7 — while the house you put it in went the other way. If you’ve ever wondered why decades of genuine progress don’t feel like progress — a major obsession of this newsletter — housing is a big part of the answer. The gains are real, but our rents and mortgages are eating them. 

That’s the bad news. Here’s the good news.

A new law for our single biggest problem

What made all this feel so depressingly unfixable is that no one seemed positioned to fix it. Zoning is controlled by thousands of city councils and planning boards, each answerable to neighbors with a vested interest in scarcity, since for most American homeowners, the housing shortage bolsters their net worth. And Congress had simply left the field, going roughly 30 years without passing a major housing law. 

Then, over about three weeks this summer, lawmakers acted. On June 22, the Senate passed the 21st Century ROAD to Housing Act 85-5. The House followed a day later, 358-32, and after President Donald Trump chose not to veto it, the bill automatically became law on July 11. The ROAD Act is the most significant housing legislation in decades, and the first built squarely on a basic YIMBY premise that cuts to the heart of the housing theory of everything: Homes are expensive because America made them too hard to build.

The act stitches together more than 60 separate bills, 36 of them bipartisan, negotiated by Sen. Tim Scott (R-SC) and Sen. Elizabeth Warren (D-MA) along with Rep. French Hill (R-AR) and Rep. Maxine Waters (D-CA). Its centerpiece, the Build Now provision, ties federal block-grant money to results: cities that add homes get more, cities that block them get less, and a $200 million annual innovation fund rewards measurable increases in supply.

The rest of the bill takes scissors to the red tape strangling housing. The law streamlines federal environmental review for housing the federal government itself helps fund. It directs the Department of Housing and Urban Development to write guidelines for single-stair apartment buildings up to six stories — a small-sounding change that my colleague Rachel Cohen Booth has called “a deceptively simple reform that could unlock more housing.” And it ends a 1970s-era rule requiring factory-built homes to sit on a permanent wheeled chassis, a mandate that added thousands of dollars per unit and helped keep the cheapest form of American housing out of most neighborhoods.

What’s revolutionary here isn’t the elements of the bill — which are still largely small-bore compared to the scale of the problem — but the acceptance of the basic YIMBY idea that, as Republican Sen. John Kennedy of Louisiana said, “If we had more housing, the price would go down.” Ben Metcalf of UC Berkeley’s Terner Center told the New York Times the law was “catch-up on 30 years of policy,” while Laura Foote, executive director of YIMBY Action, put it more simply: the law is a triumph “simply because it exists.”

The states ran the experiment first

Congress may be late to these YIMBY ideas, but it has the benefit of knowing they have been tested. One of the best examples is in Auckland, New Zealand, which upzoned three-quarters of its single-family land in 2016, rewriting the rules so those lots could hold apartments, not just one house each. Construction roughly doubled within five years, and a study published last year estimates Auckland rents are now about 23 percent lower than they would have been without the reform.

The American version of Auckland — with better tacos — is Austin, which spent a decade legalizing apartments, killing parking minimums, and shrinking minimum lot sizes. The Texas city added 120,000 homes between 2015 and 2024, growing its housing stock by 30 percent. As my colleague Marina Bolotnikova reported this spring, Austin rents fell 6 percent in a single year, more than any other large US metro, with the fastest declines in older, cheaper buildings — exactly where relief matters most. 

Other states have noticed. California exempted most urban infill housing from its famously litigious environmental review law. Montana passed its “Montana Miracle” package, legalizing duplexes and backyard apartments on land that had allowed only single houses. Texas legalized homes in commercial zones statewide. 

Red states and blue states are coming to the same diagnosis about solving the housing problem: build more. The new federal law mostly just tells them to keep going.

Why your rent won’t drop tomorrow

It’s a sign of just how terrible federal housing policy has been for so long that what is ultimately a pretty modest law is being greeted so rapturously. 

The ROAD Act notably contains almost no new money — its final section is literally titled “No Additional Funds Authorized.” Zoning, which can make or break housing, remains a local power. The law mandates nothing; it nudges with grant formulas, which a determined city council can shrug off. The New York Times’ Conor Dougherty, one of the top reporters on housing, judged it “unlikely to do much to blunt the high cost of rent and ownership in America anytime soon.” Mortgage rates have been stuck above 6 percent since 2022 — as I could show you from my own mortgage statements. National homebuilding has barely moved, and forecasters expect little change this year.

Experience shows that a problem big enough to be a viable theory of everything is going to need more than one reform. Minneapolis famously ended single-family zoning in 2018, and only got a modest building response, because a dozen other rules still stood in the way. What worked in Auckland and Austin was the full stack — density plus permitting plus parking plus lot sizes. Housing, as advocates put it, is a door with many deadbolts; this law unlocks the federal ones and hands the states a better set of keys. The others are still bolted.

Usually with this newsletter I like to point at progress that has already arrived but gone unnoticed. Housing is the opposite case: The problem is exactly as bad as everyone feels it is, and what arrived this month is agreement about why. America spent 40 years treating expensive housing like the weather: unfortunate, unchangeable, nobody’s fault. It has taken five years for that diagnosis published on a magazine website to become a universally praised act of Congress.

The ROAD Act won’t pour a single foundation. What it did do was build the consensus upon which the next few million of them will be built. And consensus, in American politics, is the material that takes the longest to set.

A version of this story originally appeared in the Good News newsletter. Sign up here!

Scientists have started calling it the “Godzilla El Niño.”

The nickname comes from actual climate researchers, which should tell you something; they are not generally in the habit of comparing climatic systems to mega-monsters. What’s building in the eastern Pacific right now is a surge of ocean warming that is likely to make 2027 the hottest year in recorded history, possibly by a lot. The median projection puts next year at about 1.7 degrees Celsius above the preindustrial average. The high end gets close to 1.9, which, until recently, was a number climate models saved for the late 2030s.

David Wallace-Wells has spent a decade thinking about what numbers like that mean for the people who have to live under them, first in his 2017 essay “The Uninhabitable Earth” and the 2019 book that grew out of it, now as a writer and reporter at the New York Times. In a recent piece for the New York Times, Wallace-Wells made the case that 2027 could work as something stranger than a disaster forecast: a preview of everyday life in about 2035. It could be a full year lived at a temperature we weren’t supposed to reach for another decade: fire seasons, atmospheric rivers, disrupted monsoons, strained food systems, and all. 

What’s less certain is if any of this will change our politics. Americans tell pollsters they’re more worried about climate change than ever — more than after Katrina, more than at any point in the Bush, Obama, or Biden years — even as elite discourse on the subject has been waning. Meanwhile, the green transition keeps compounding. More than 90 percent of the new energy capacity built in America last year was green — and under an administration doing its best to stop it.

Sean Illing talked through all of it with Wallace-Wells for this week’s episode of The Gray Area. When Sean asked whether 2027 will be a real test of how much climate disruption we can manage, Wallace-Wells doesn’t blink: “I think it’s going to be a test on that score, which we will fail.” 

So yes, it gets dark. But, it also ends with the most convincing case for hope on climate I’ve heard in a while — and I write a newsletter that goes looking for good news every week. As always, there’s much more in the full podcast, so listen and follow The Gray Area on Apple Podcasts, Spotify, Pandora, or wherever you find podcasts.

This conversation has been edited for length and clarity.

Just about everybody has heard the term El Niño, but most of us — myself included, until very recently — don’t really know what it is. So let’s just start there. What is an El Niño?

It’s a climatic event that happens irregularly but pretty often — usually about every three or four years, sometimes as infrequently as every seven — in which you get a burst of warmer-than-normal temperatures in the coastal Pacific, off the coast of South America. And that has a lot of amazingly global consequences, in ways that are not intuitive to most normies.

So, in this case, you tend to see a lot more rain across the western US. You tend to see a lot less rain across Brazil and the Amazon. You see disruption to the monsoon season in South Asia; a similar disruption to rain across the Sahel and sub-Saharan Africa; and, typically, a suppression of hurricane activity in the Caribbean. Which is all to say: It’s a global disruption to the basic normal patterns that we come to depend on, but the disruption isn’t all in one direction.

At the global level, it makes things hotter — and the one that we’re seeing forming this year may be a record-setting one. I think the best way to think about it is just like we shook up the whole planet’s climate system, and we see a lot more disorder than we see in non-El Niño years.

What is particularly scary about this El Niño? Why do scientists seem to be alarmed?

There are a few different components to it. The first is that we are going to experience it at a much higher baseline of global average temperature than has ever been experienced before. Given that we are warming the planet quite rapidly, just about every year is warmer than just about every year that came before. We’ve normalized it so much that we don’t raise our eyebrows when we set new records for global average temperature. But novel events like an El Niño are happening on top of that baseline.

The things that terrified us into action not that long ago are now our inevitable future.

But another reason why scientists are concerned is because it is a super El Niño. Some of them are even calling it a Godzilla El Niño, which is to say the boost that we’re going to see because of this event may be bigger than any similar event ever observed in modern history.

That’s particularly alarming, because the El Niños of the late 1870s produced perhaps the most human suffering that the modern world has ever experienced: famines that killed tens of millions of people all around the world, from Africa to South Asia to China. I don’t want to suggest that this El Niño is capable of producing that kind of suffering. Obviously, 2026 is a different world than 1878, but it gives you a framework for how the consequences of an El Niño may filter through our political world if, over the next year, we see significant crop failures in multiple parts of the world, which is maybe even more possible because there’s a fertilizer shortage as a result of the war in Iran.

We probably will be able to respond to that such that 10 or 50 million people don’t die, but there will be considerable political disarray that results: bread riots, political protests, civil unrest, perhaps in multiple parts of the world. I’m not saying that’s inevitable, but it reminds us that a huge amount of what we take for granted as stable features of modern life is nevertheless subject on some level to disturbance by the climate system. And this climate system this year is winding up to really give us quite a wallop.

You write that next year, 2027,could be a preview of what everyday life will be like in a decade or so. Talk about that.

So, here we are in 2026, and the global average temperature is about 1.5°C above the preindustrial average. It doesn’t sound like very much, and on any given day that kind of effect is basically imperceptible to the average person. But it is a global average, and it hides a lot of extremes.

Our elite discourse and our elite policymaking has somehow decided to treat those anxieties as irrelevant.

It is also right about what we agreed we would try to limit warming to in the Paris agreement in 2015. We said we would definitely keep warming below 2 degrees, and we would do everything we could to keep warming close to 1.5 — a number that came from emerging science and emerging political sense among the world’s most vulnerable countries that 2 degrees wasn’t enough. That 1.5 degrees was itself going to be quite difficult to manage, but anything north of that was going to be quite catastrophic.

When we signed that agreement just 10 years ago, we were not yet north of 1 degree of warming. We are now, in the long-term average, somewhere about 1.3 or 1.4, [with] a single year at 1.5. And we are heading quite rapidly not just past the more ambitious goal we set for ourselves but toward the less ambitious one: 2 degrees. It now seems, practically speaking, inevitable. We’re heading toward a future that we defined as unacceptable.

And when I say that, I don’t just mean that the scientists at the IPCC defined it as unacceptable. When they published their research into the difference between those two levels that report was the event that gave us the huge global climate awakening of late 2018, 2019, into 2020. It’s the reason we know who Greta Thunberg is. It’s the reason we heard of Extinction Rebellion and Sunrise. It’s the reason we started talking about a Green New Deal. All the lip service that presidents, and prime ministers, and CEOs paid to climate change in the years that followed came out of a report that was basically saying, “We need to limit warming below 2 degrees.”

And we now know that we will be heading north of 2 degrees. The things that terrified us into action not that long ago are now our inevitable future.

Is it your belief that 2027 will be a real no-bullshit test of how much climate disruption we can effectively manage at this stage?

I think it’s going to be a test on that score, which we will fail. And the question is what impact that makes socially and politically. We are already seeing every year, in countless ways, how poorly prepared we are for the climate present, but most of the time it isn’t producing the kind of political and social response that I would like to see. And I worry that the basic pattern is going to continue, even into a much warmer future, where, rather than changing course and designing a much more responsive and resilient future, we will just find ourselves defining more and more suffering as acceptable and normal as the cost of going about our daily lives without being too preoccupied by the course of climate change.

A decade ago, a lot of people — myself included, I guess — really did believe that extreme weather would eventually force climate action. Can we just say by now that that theory has failed?

People are complicated, populations are complicated, and politics is complicated. I think of this story going back to Katrina, which was an eye-opening event more than 20 years ago now. It seemed, for a time, to mark a high point in American climate concern. It helped discredit the presidency of George W. Bush, gave us an upsurge of climate urgency and the election of Barack Obama — who said this would be the moment we looked back on as the moment the seas slowed their rise and the planet began to heal.

“We have imposed on ourselves and our future much more suffering and difficulty than we needed to, because climate change will be that disruptive.”

We’re going to be muddling through, and the future that we live in as a result will have a lot of bad stuff in it and a lot of good stuff in it.

We’ve seen a bunch of climate disasters just in the United States since then. Again and again, there’s a period of intense national focus and, then, a kind of withdrawing from the story of climate change. But that narrative — that we keep getting knocked back on our heels by climate disasters and then forget about them entirely — is also too simplistic. Looking at the headlines, you might think that everybody moved on from climate. The truth is that when you look into the survey data, we’re basically at the highest point for climate anxiety that we’ve ever been. Americans are more concerned about climate change than they were in the aftermath of Katrina. They’re more worried than they were at any point in the George W. Bush presidency, the Barack Obama presidency, and the Joe Biden presidency.

So, we have an interesting political story here. We’re telling ourselves that the country has moved on from climate. But, in fact, at the level of individual people — how they feel about their lives, the lives of their communities, and the future of their country — they are just about, if not more, worried about climate impacts than they’ve ever been before. And our elite discourse and our elite policymaking has somehow decided to treat those anxieties as irrelevant.

But the question is not whether they’re worried. The question is whether they’re worried enough to actually sacrifice anything in order to make headway.

You often hear from people who think we needn’t worry as much about climate change that most Americans are not willing to pay, say, $10 a month to take action on climate. In some ways, I feel like that’s a way of weaponizing American entitlement against one particular set of policy crusades. But the other thing I would say is that it’s not the case that we need to shoulder burdens in any meaningful sense to take action.

Solar power is dramatically cheaper than any fossil fuel alternative that’s ever been produced — at least by the levelized cost of electricity, which has been the industry standard for a generation. That measure is misleading, because it only accounts for the marginal cost of electricity and not the overall cost of a system; you get more solar when the sun’s shining and more wind when the wind’s blowing, so you have to build out around that to have a reliable supply. But because batteries have now made the same dramatic improvement in their cost that solar did a decade ago, you can use very cheap batteries to even out the distribution of solar and wind power. And that means the overall cost, according to recent research, of renewable power is now lower than any alternative system. We know those price declines are going to continue for renewables, and we know that they will not decline for fossil fuels.

I’m not an economic determinist — politics plays a huge role here, as does culture — but look at the United States right now. We have a president in the White House who is remarkably opposed to renewable energy. He has done an awful lot to try to kneecap the development and rollout of renewables in America. And, yet, last year, when he returned to office, more than 90 percent of all the new energy infrastructure built in America was green. For every unit of new fossil fuel infrastructure we were building, we built nine times as much green energy. This year, the share is going to be even bigger. It’s expected to be 93 percent, which is about what it is around the world entirely.

And with the Iran war, we see many countries around the world looking at the logic of their dependence on fossil fuels and thinking, “Why would I depend on international actors to continuously supply a power source at prices that they determine when I can do a one-time importation of green energy infrastructure and then collect the dividends in the form of sun and wind, which never stop and are domestically controlled?” If you’re talking about this as a race, it seems obvious that the green stuff is winning. The problem is that it’s not winning nearly fast enough to avoid some of these quite scary climate outcomes.

David, I feel smarter at the end of this conversation, because I know more, but I do not feel more optimistic about tomorrow. Do you have any good news that we can dip out of here with?

The world is full of good news. I can be the famous optimist David Wallace-Wells. Just on climate stuff: The green transition is proceeding at a breakneck pace, much faster than even its biggest boosters said was possible 10 or 15 years ago. The battery revolution changing the economic calculus for renewables is really profound. It means that we can probably, in large parts of the world, do 100 percent renewable power, which many people said was going to be completely impossible. And even before we get there, we are already dealing with free electricity in the middle of the day in sun-rich parts of the world. In Australia, they’ve literally passed a law promising citizens that, for several hours every day, they will be guaranteed free electricity.

And that’s just power. With the EV rollout that we’re seeing all around the world, you never have to go to a gas station ever again, because you just charge up at your house. Whenever we electrify anything, it gets dramatically more efficient. And we’re cleaning up air pollution, which is a much larger threat to human flourishing on any conceivable timescale than climate change is. Today, somewhere between 5 and 10 million people are dying every year from air pollution, but we’re past the peak of that. We’re drawing it down. There’s a lot of progress that is really wonderful and worth celebrating.

So, I don’t want to give the sense that, for all of my political despair, and for all of what I would call my climate realism, we are all doomed in some meaningful way. We’re living in an incredibly messy world. We have imposed on ourselves and our future much more suffering and difficulty than we needed to, because climate change will be that disruptive. But that challenge is a political one, it’s a social one, it’s a technological one. And while we are not pursuing each of those paths optimally, in my view, we also shouldn’t assume that we will fail each of those tests like we’re getting a zero.

We’re going to be muddling through, and the future that we live in as a result will have a lot of bad stuff in it and a lot of good stuff in it, which is, of course, human nature and human history. One hopes along the way we will wake up a little bit more — not just on climate, but on politics and governance. My hope, when I’m feeling optimistic, is that these stories are playing out on long enough timelines that we have at least the opportunity to take control of our climate system, our economic system, and our political system in ways that have eluded us in the past.

Listen to the rest of the conversation and be sure to follow The Gray Area on Apple Podcasts, Spotify, Pandora, or wherever you find podcasts.

It’s been feeling very 1995 lately. 

Oasis, whose last good song came out back when Macs were still see-through, was the second-biggest tour on the planet last year. Gen Z is snapping up the flip phones and digital cameras their parents had traded in for smartphones, and Y2K-era fashion is flooding the mall, in case you were wondering where all the rhinestones came from. Pop star Olivia Rodrigo just released a hit album heavily influenced by alternative rock recorded a decade before she was born. 

Ask Americans directly and they’ll tell you: the 1990s are the decade we’re most nostalgic for, edging out even the boomer-beloved ’80s. 

Speaking as someone who lived through the 1990s as a teenager and mostly enjoyed it (at least in retrospect), it makes total sense to me. Of course, it would — as a Washington Post analysis found, when Americans name the country’s “best decade,” the answer inevitably boils down to whichever decade they turned 11. 

But personal experience also means that, with a little effort, I can remember what the 1990s were actually like to live through, and the reality was a different story than the myth — especially for teenagers of that era.

Take teen pregnancy. In 1991, the teen birth rate hit a record high of 61.8 births per 1,000 girls aged 15-19. In his 1995 State of the Union address, President Bill Clinton called “the epidemic of teen pregnancies and births where there is no marriage” America’s “most serious social problem.” A National Campaign to Prevent Teen Pregnancy formed the following year in direct response. 

And the fear ran well beyond pregnancy: juvenile arrests peaked in 1996 at nearly 2.7 million, at the height of the so-called “superpredator” panic over a coming wave of remorseless teenage criminals. In 1995, as the writer Adam Mastroianni points out in an excellent new essay that helped inspire this piece, half of high school students drank alcohol, 35 percent smoked cigarettes, 40 percent had at least tried marijuana, one in 10 had brought a weapon to school at some point, and about 6 percent of teenage girls were pregnant in any given year. The teen suicide rate for boys hit a record high in 1990. Americans were terrified for their teenagers, and terrified of them. 

So, no, the 1990s weren’t all teen boy bands, high school comedies, and endless economic growth. But what’s amazing, at a moment when everyone seems sure that kids today are not all right, is just how much those scary signs of teen dysfunctions from the 1990s have simply vanished.

This spring, the CDC reported that the US teen birth rate fell to 11.7 births per 1,000 girls ages 15 to 19 in 2025 — down 81 percent from its 1991 peak, and the lowest figure ever recorded. That has been the pattern for more than three decades now: the teen birth rate sets a record low, then breaks it, then breaks it again.

It’s not just teen pregnancy: nearly every risky teen behavior of the halcyon 1990s started collapsing almost at once, kept collapsing for 30 years, and is collapsing still. The teen birth rate dropped 7 percent in 2025 alone. And yet it’s hard to find anyone who planned this victory, and nearly impossible to find anyone who celebrated it.

The kids got careful

Let’s start with teen births, which largely comes down to two factors working in concert: fewer teens are having sex, and those who are use much better contraception. 

The share of high schoolers who have ever had sex fell from 54 percent in 1991 to 32 percent in 2023, and Guttmacher Institute researchers found that improved contraceptive use drove essentially all of the pregnancy decline in the late 2000s and early 2010s, including a sevenfold rise in teens using long-acting IUDs and implants. In the decade that followed, delaying sex became the biggest single contributor. 

But contraception can’t explain why violence fell too. The best unified theory comes from psychologists like Jean Twenge and more recently writers like Mastroianni, whose “Decline of Deviance” essays argue that kids stopped taking risks of all kinds in part because their futures became too valuable to gamble. As lives got longer, safer, and richer, the expected cost of any given joint, fight, or pregnancy went up. Twenge called this the “slow life strategy,” and it shows up as much in data about taking good risks as it does in taking bad ones: the share of high school seniors with a driver’s license, for instance, fell from 82 percent in 2005 to 63 percent in 2023.

If you’re waiting for smartphones to enter the picture, you’re going to keep waiting — most of these trends began in the 1990s, long before broadband, let alone Instagram. And policy didn’t seem to do much either. England spent hundreds of millions of dollars on a national teen pregnancy strategy, and a comparative analysis found rates fell about as fast as they did in similar countries that spent nothing. (Researchers have found that one of the few interventions with credible causal evidence behind it, at least in the US, was MTV’s iconic 16 and Pregnant — a reality show that worked by making teen parenthood look as exhausting as it was in reality.) 

Everything everywhere at once

The same collapse in dangerous behavior shows up everywhere you look. Half of high school students drank alcohol in 1991; 22 percent did in 2023. Seventy percent had tried a cigarette in 1991; under 18 percent had by 2021 — and no, they didn’t all just switch to vaping. E-cigarette use has also been falling since 2019. The share of high schoolers who’d been in a physical fight in the past year dropped from 42.5 percent in 1991 to 22 percent by 2019. And this year’s Monitoring the Future survey found that 66 percent of 12th graders — a record — used no alcohol, marijuana, or nicotine at all in the past month.

For me, no number stands out more than marijuana. When I was a teen in the 1990s… well, when I was a teen in the 1990s I was way too uncool to know anyone who had marijuana. (Honest, Mom.) But weed is now legal for adults in half the country, almost certainly easier for a teenager to obtain than at any point in American history, and generally regarded by young people as less dangerous than their parents ever believed. Yet even as marijuana use has spiked among adults, it has fallen among teens from its late-1990s peak. 

All that caution shows up in the most basic stat there is: teenagers stopped dying so frequently. In 1990, 88 of every 100,000 Americans ages 15 to 19 died each year; by 2013 it was 45. The single biggest reason was the car: since 1975, teen road deaths have fallen 67 percent thanks to safer vehicles, the graduated licensing laws every state adopted starting in the mid-1990s, and all those teens who never bothered getting a license in the first place. At the 2013 low, an American teenager was half as likely to die as one in 1990. 

That last stat is worth holding onto. Nostalgia isn’t just misleading; it also suffers from survivorship bias. Adults like myself may look back on their 1990s adolescence as a golden age, but there are tens of thousands of my peers who never lived to look back.

The price of caution

The problem with reducing risk-taking behavior across the board is that not all risks are bad. The same caution that might keep a 16-year-old from smoking or having sex might later keep her from founding a company, moving across the country, or writing a strange novel. In Mastroianni’s “Decline of Deviance” essays he documents falling cultural and scientific risk-taking right alongside fewer fistfights. “For the first time in history,” he writes, “weirdness is a choice. And it’s a hard one, because we have more to lose than ever.””

And if teens are so safe, why are they so sad? Forty percent of high schoolers reported persistent sadness or hopelessness in 2023, and the social psychologist Jonathan Haidt has argued that we’ve overprotected kids in the physical world while underprotecting them in online spaces that didn’t even exist in the dial-up 1990s. The timing backs him up: when the CDC first asked the sadness question in 1999, about 28 percent of students said yes, and the number barely moved for nearly two decades until surging after 2017. Physical safety and psychological well-being may turn out to be more separate than we ever realized.

But our valid worries about the mental state and uncertain future of today’s teens shouldn’t blind us to just how remarkable the shift from the 1990s is as a whole. As Mastroianni writes, “We would have spent billions to solve all of these problems back in the 1990s… and yet when we got the thing we wanted so badly, we didn’t even notice.” 

America wanted its teenagers to stop getting pregnant, stop smoking, and stop fighting — wanted it badly enough to put it at the center of a State of the Union address — and then got nearly all of it, mostly free of charge, from the kids themselves.

“Can we take a win?” Mastroianni asks. We can start by counting this one.

A version of this story originally appeared in the Good News newsletter. Sign up here!

This story was originally published on June 29 in The Highlight. To get access to member-exclusive stories like this every month, become a Vox Member today.

America in the summer of 1976 was not in a good place. 

The president who presided over the country’s bicentennial, President Gerald Ford, only had the job because the previous president and vice president had resigned in disgrace, making him the sole US president who was never actually elected. The Vietnam War had ended in defeat and disgrace when Saigon fell the year before, after the deaths of nearly 60,000 American servicemembers. Inflation hit double digits in 1974 and stayed ugly, unemployment sat near 8 percent, and economists had to invent a word — stagflation — for an economy that seemed to encompass the worst of both worlds.

Given all that, you might assume the national mood leading up to the 200th anniversary was grim. And, yet, on July 4, 1976, something strange happened: Americans threw themselves a hell of a party. 

In New York Harbor, more than 200 tall ships sailed up the Hudson for Operation Sail, drawing an estimated six million spectators — the largest crowd in the city’s history. Ford reviewed the fleet from the deck of the aircraft carrier USS Forrestal. It was the same scene up and down the country that day: parades in small towns, fireworks over the National Mall, church bells ringing in unison at 2 o’clock. It was one cathartic day of celebration after a decade that had offered little reason for it.

And when pollsters asked people how they felt about the country’s future that year, the mood was, improbably, sunny. A Roper survey found more Americans were optimistic than pessimistic about the future by a nearly three to one ratio. More than three-quarters told Gallup the nation had already achieved at least a fair amount of its founding ideals. Somehow, a nation that was in the middle of a genuinely miserable decade looked in the mirror and liked what it saw.

Jump forward 50 years, to this year’s 250th anniversary, and you’ll find the vibes flipped. Roughly 60 percent of Americans tell pollsters the nation is on the wrong track. A majority say its best years are behind it. About three-quarters think today’s children will end up worse off than their parents. Asked a version of that same founding-ideals question from 1976, 77 percent now say the founders would be disappointed in what we’ve become.

But just as they were in 1976, the vibes don’t match reality. Set the mood aside and look only at the numbers, and the country that felt so good in 1976 was, by the most important measures, a worse place to be alive than the country that now feels so terrible on its 250th birthday.

Start with whether you’re alive

Let’s start with the most basic test of how a society is doing: how long its people live.

Life expectancy at birth in the US was 72.6 years in 1976. In 2024, it reached a record high of 79 years — an extra six and a half years of life. At the start of life, a baby born now is far more likely to survive its first year than one born during the Bicentennial, while cancer, once nearly a synonym for a death sentence, now kills a much smaller share of the people it strikes. 

The US made those gains by stopping some of its worst habits, things that were commonplace in 1976 . You might have seen the Bicentennial celebrations through a cloud of smoke, as cigarettes were woven into ordinary life — on airplanes, in offices, in hospital wards — and roughly 37 percent of adults smoked. Today, it is closer to one in 10, and it keeps falling. 

The heart disease and lung cancer that were connected to all that tobacco have receded with it. Add seatbelts and airbags, better trauma care, and cheap drugs that lower cholesterol and blood pressure, and the result is a country where the things that were most likely to kill an American in 1976 are less deadly now.

The America of 1976 sat at the leading edge of a brutal crime wave; the murder rate would peak in 1980 and stay high for more than a decade. By the early 2020s, however, violent crime had fallen back to roughly a 50-year low, and homicide rates this year may end up at a record low. And the single most dangerous thing most Americans do — get behind the wheel of a car — is far less likely to kill them, with the death rate per mile driven now a fraction of what it was at the Bicentennial.

The country got cleaner, and richer, and fairer

In 1976, the air in American cities carried lead, an honest-to-God neurotoxin that was pumped out of every tailpipe of the more than 90 percent of American vehicles that used leaded gasoline. 

Rivers literally caught fire: The Cuyahoga in Cleveland had burned so many times it became a national joke, and Lake Erie was widely written off as dead. And things were bad outside Ohio, too. In Los Angeles, the smog got thick enough to keep kids inside at recess and erase the nearby mountains from view.

Since 1970, however, the combined emissions of the six main air pollutants the EPA tracks have fallen 78 percent — even as the economy nearly quadrupled in real terms, the population grew by tens of millions, and Americans drove far more miles. That split, with growth going one way and pollution the other, is one of the least celebrated but most consequential triumphs of the past half-century, the product of legislative efforts and technological response. And lead? It’s essentially disappeared from the air. 

And it’s not just economic or environmental statistics that have improved; society advanced, as well. Women now earn the majority of college degrees. The Black poverty rate sits near a record low. Support for same-sex marriage is now the norm — maybe the single biggest social change from 1976, when homosexuality was criminalized in most states. Pick a metric more or less at random, and the line usually runs the right way.

This is not a matter of cherry-picking a few flattering numbers. It is the overwhelming direction of the evidence, across health, wealth, safety, rights, even the basic cleanliness of the physical world an American walks through every day. Measured against its own recent past, the US is in some of the best shape it has ever been.

So what’s with the bad vibes?

A more perfect union doesn’t mean perfect

Well, some things genuinely got worse, and they are not insignificant. 

Americans’ faith in their government has collapsed; fewer than one in five now trust Washington to do the right thing, down from solid majorities in the 1960s — and the country is more polarized than it was in 1976. Democratic decline and even collapse is a live threat. Those economic gains I highlighted above have flowed disproportionately upward. The top 1 percent’s share of income, near a historic low in 1976, has since roughly doubled.

Climate change barely registered in 1976. The carbon dioxide in the atmosphere has since climbed from around 330 parts per million to about 427, and warming will only get worse in the future. And buying a home increasingly feels out of reach for many. By 2024, a record share of households spent more than a third of their income on housing. (Notably, though, the percentage of Americans who own a home is slightly higher than it was in 1976, and those homes are much larger on average.) 

These are real problems, but they remain exceptions to a broader half-century trend of improvement. And a country that scrubbed the lead from its air and put out smoking can overcome new challenges, as well. 

Which brings us back to a tale of two birthdays. In 1976, Americans had less of nearly everything you can count, and, yet, they reported feeling good about the future anyway. In 2026, we have more, and we don’t. 

Just as it can be for a person, a country’s mood is a poor instrument; it measures the story we are telling ourselves more than the lives we are actually living. For all our pessimism about the state of the nation, more than three-quarters of Americans say they are satisfied with their own lives.

The Americans crowding New York Harbor in 1976 were cheering a country that was sicker, dirtier, more dangerous, and less free than the one we live in now. But they were right to cheer; the line was already bending the right way, and it kept bending. It turns out a nation can travel a long way, even while it is convinced it is going nowhere.

A version of this story originally appeared in the Good News newsletter. Sign up here!