This is part two in our three-part series on space. We explore how the modern commercial space age came to be — both the context and forces driving toward the creation, quite literally, of a highway to the heavens. If you missed it, here’s part one.

Solving For takes on one hard problem at a time — unpacking the stakes, exploring the forces behind it, and surfacing real paths forward. Each series unfolds in weekly installments. You can read or listen to each series — narrated by me — at solvingfor.io, or click the article voiceover at the top of this page.

In 2008 SpaceX proved that a private company could put a rocket into orbit. Then, across three consecutive years, from 2015 through 2017, it proved that a rocket could be brought back to Earth and flown again.

The first was a proof of concept. The next three, taken together, amounted to the hinge moment that brought today’s commercial space age.

SpaceX’s 2008 launch from a remote outpost in the Marshall Islands was a watershed. After three failed attempts, Falcon 1 reached orbit, showing that launching rockets wasn’t just the domain of governments and their traditional contractors. Entrepreneurs and engineers could do it too.

But the breakthrough that best explains the present — when we see multiple launches in a week, thousands of satellites overhead, a private operator landing a spacecraft on the Moon, and credible ambitions for a lasting foothold on the Moon and, eventually, Mars — started with a booster rocket falling out of the night sky.

It was on December 21, 2015 that a SpaceX Falcon 9 lifted off from Cape Canaveral carrying 11 satellites into orbit. Minutes into its flight, the booster rocket separated and began falling back toward Earth. But instead of crashing into the ocean, it reignited its engines, slowed its descent, and touched down upright a few miles from where it had lifted off. (See the video.)

It was the first time an orbital-class rocket had ever landed intact.

Subscribe now

It was a big deal. A booster rocket accounts for up to 70 percent of the cost of an entire rocket. If it could be reused, the arithmetic of reaching space would change dramatically.

But landing was only the first step. Next came an even harder mission: landing a rocket on a barge at sea, which is critical for boosters with too little fuel to return to the launch site, and then fly it again.

On April 8, 2016, a Falcon 9 delivering supplies to the International Space Station sent its booster descending toward Earth. The booster touched down, landing upright, on a 170-by-300 foot barge called “Of course I Still Love You,” nearly 200 miles off the Florida coast. (See the video.) The name comes from an Iain M. Banks science fiction novel.

Then, on March 30, 2017, the final piece of the puzzle: the same booster that had landed on the barge the year before was launched again, this time carrying a communications satellite into space. (See the video.) Once again, the booster rocket landed upright on the “Of course I Still Love you” barge.

Land a booster rocket. Recover it. Fly it again.

Reusability had been unlocked, and the cost of going to space was about to drop dramatically.

We Chose To Go

This feat, of course, stood on the shoulders of nearly sixty years of the U.S. space program. NASA was started in 1958, a year after the Soviet Union shocked the world by putting the first satellite — Sputnik — into orbit. To the U.S. this wasn’t a scientific achievement; it was a warning shot.

First came NASA’s Mercury program, then Gemini. John Glenn orbited Earth three times in February 1962. Three years later astronauts were docking spacecraft and walking in space. In between, John F. Kennedy stood at Rice University and gave the program its purpose: “We choose to go to the moon in this decade and do the other things,” he declared, “not because they are easy, but because they are hard.”

Then came the Apollo program. From July 1969 to December 1972, the U.S. made six successful landings on the moon, resulting in 12 astronauts walking on the lunar surface.

It began with Neil Armstrong stepping onto the moon’s Sea of Tranquility on July 20, 1969. It ended with humanity’s longest stay on the moon to date. Eugene Cernan and Harrison Schmitt spent three days on the moon, driving a rover through the moon’s Taurus-Littrow Valley, venturing miles away from their spacecraft on the rocky lunar landscape, and collecting the largest haul of lunar samples any crew has ever brought home.

When Cernan climbed the ladder on December 14, 1972 to fly home, he became the last human to stand on the Moon.

Share

Each of the missions had one funder and one customer — the U.S. government — and essentially one goal: to beat the Soviets and do it in a public way for the world to see.

The program built the ability to go to the moon seemingly whenever it wished — the only blemish on a trip to the Moon being when the Apollo 13 mission had to abort; astronaut James Lovell famously guided the troubled spacecraft back to Earth.

Yet, after achieving six moon landings in three and a half years, the U.S. lost interest. The final three planned Apollo missions were canceled. People had moved on. No human has set foot on the Moon again.

Stubborn Costs

The Space Shuttle program would come soon after and it was supposed to fix the Apollo program’s biggest flaw: namely, the high costs. The Saturn V rocket, which took astronauts to the moon, had a payload cost of about $5,400 a kilogram to reach orbit, according to a Center for Strategic and International Studies report. NASA originally pitched the Shuttle as something that would be much cheaper.

The Space Shuttle program flew 135 missions, conducted cutting-edge science in orbit, and built the largest structure in space, the International Space Station. But instead of costs going down, they went up — a lot. When the program’s full costs were tallied, the real number came out to roughly $65,400 a kilogram, a dramatically higher cost than the rockets it was meant to replace.

NASA and its longtime contractors — Lockheed Martin, Rockwell International, McDonnell Douglas — didn’t break the costly cycle. By the mid-2010s, the inflation adjusted cost for payloads to low Earth orbit was about $13,600 a kilogram, better than the space shuttle but worse than the days of the Apollo program.

These high costs are why so many of the things explored in Part 1 didn’t exist. A satellite constellation that needs thousands of launches was prohibitively expensive. So was a data center in orbit, or manufacturing pharmaceuticals in microgravity.

Space stayed where it had been: a government project that did much for science, defense and prestige, but not a place to run a business.

Breaking the Cycle

SpaceX’s successful launch in 2008 showed what was possible by putting a privately-funded, lower cost rocket into orbit. And from 2015 through 2017, it showed this could be done at scale by reusing the most costly part of a rocket, the booster.

But it needed to break the high costs that made going to space prohibitive for even most governments. Broadly speaking, SpaceX did it by doing three things.

Subscribe now

One, first principles and vertical integration. Rather than buying parts from a supplier, Musk explored what raw materials are needed to build a part and how much it would cost, versus the price of a manufactured part. He called this the “idiot index.” For instance, if a metal bracket costs $1,000 but the raw aluminum costs only $100, the idiot index is ten. When that number was especially high, he would build it from scratch. In time, he created the capability at SpaceX to build many rocket parts in-house.

Two, rapid iteration and failure. In 2012 and 2013 SpaceX began testing a rocket called the Grasshopper that would climb thousands of feet into the air and come back down, landing upright. Each test flight climbed a little higher than the last. Eventually came attempts at landing on floating barges. The more ambitious the test, the more frequent the crashes. (Watch a SpaceX crash video.) But the learnings came just as fast. By 2017 complete success was achieved.

Third, the now-famous process that Musk calls: “the Algorithm.” This is the five step process used at both SpaceX and Tesla:

• Make requirements less dumb. Each requirement should come with the name of the person who made it. Each requirement should be questioned, no matter who made it.

• Delete the part or process step. Aggressively edit out steps in any process. A report by SpaceX investor Andreessen Horowitz noted that if you do not end up adding back to at least 10 percent of what you deleted, “you did not delete enough.”

• Simplify and optimize. This comes after step two because you don’t want to simplify or optimize a step that shouldn’t have existed in the first place.

• Accelerate. After doing the first three, find ways to speed things up.

• Automate. With everything in place and the bugs shaken out, automate everything that’s repeatable. Many organizations move to this step without the first four.

The result is that SpaceX cut the cost of launching a rocket into orbit by more than 95 percent, as compared to the Space Shuttle, according to the CSIS report. The payload cost went from $65,400 a kilogram in the Space Shuttle era to around $2,600 a kilogram on SpaceX’s Falcon 9 rocket. And for its Falcon Heavy booster launching massive payloads into orbit, the cost is $1,500 a kilogram.

Not Alone

Of course, it’s tempting to view this solely as a story of entrepreneurs disrupting a calcified system. But it’s important to note that SpaceX wouldn’t be the company it is today without NASA. The reason: NASA funded SpaceX.

One example: in 2008, months after SpaceX put its first rocket into orbit, NASA awarded the company a $1.6 billion contract to send rockets to the International Space Station on resupply missions.

“NASA, and specifically the initial commercial cargo contract, is what saved the company when it was on the brink of bankruptcy,” Chris Quilty, who leads a space-focused research firm, told CNBC in 2025.

With such contracts keeping it in business, SpaceX was able to drive down costs so it became a preferred launch provider for anyone wanting to put anything in space. In 2025, SpaceX conducted 165 orbital launches, more than the rest of the world’s launch providers combined, plus five additional test flights with its newer rocket, Starship. (SpaceX has a total of 706 orbital launches as of August 13, 2026; and is growing at a pace of two or more a week.)

And now, with costs much lower, SpaceX is building additional businesses on top of its launch business. In 2019 it began launching its own satellites, called Starlink, into space to offer internet around the world. That business now has some 10,200 broadband satellites orbiting Earth and more than 12 million subscribers in 167 countries, according to its most recent earnings report. And, it hopes, to soon put its own AI data centers, called Starmind, into space too.

Meanwhile, even as SpaceX has become so dominant, it’s not alone in opening access to space.

Exhibit A is Rocket Lab, a company founded in New Zealand by a self-taught engineer named Peter Beck. The company put a rocket into orbit in 2018. Its rocket, called Electron, targets the small-satellite market (as opposed to SpaceX carrying bigger payloads) and has become the most flown small launch vehicle in the world. To date, it’s had more than 90 successful missions to orbit.

The likes of Astra and Firefly Aerospace are in the rocket launch business as well, with Astra putting two rockets into orbit and Firefly putting three into orbit successfully. In March 2025 Firefly did the feat of putting a lunar lander on the Moon.

Meanwhile, SpaceX took another step reusing rockets faster and more cheaply. On Oct. 13, 2024 it caught its Super Heavy booster — which powers its new Starship rocket — out of the air. Instead of landing on land or a barge, it was caught by two mechanical arms — called “chopsticks” — on the launch tower itself. (See video below.) With this, there is no towing a booster back from sea or trucking it from a landing pad. It returns to the tower that it launched from, ready to be inspected and flown again.

It’s the same idea that started with the Falcon 9 in 2015 — land it, recover it, fly it again — but pushed one step further by catching it before it even touches the ground.

Ultimately, the plan is spaceflight that’s not dramatically different from commercial aviation. In SpaceX’s case that means going from twice a week to daily — or more. In February, Musk put a number to it, writing on X: “it will get really nutty when Starship is launching every hour in 3 years.”

The highway to the heavens is quite literally being built. The question is what to do with it — and what could go wrong. That’s next.

Prefer to listen? I narrate each edition myself. Scroll up to find the audio version at the top of this page.

Previous Series —

China’s Rare Earth Dominance | AI Safety | Decline of Local News | End of Amateurism in College Sports | Shrinking Competition in Congress | Social Media and Teen Mental Health | A World Rearming as the Global Rules-Based Order Weakens | America’s National Debt Crisis | Reinventing the American Dream

Solving For takes on one hard problem at a time — unpacking the stakes, exploring the forces behind it, and surfacing real paths forward. Each series unfolds in weekly installments. Learn more.

This week we launch a new series on space.

(If you’re new here, Solving For explores one pressing challenge at a time — the stakes, the forces driving it, and what a path forward might look like.)

The modern space race is accelerating fast. Rocket launches used to be major events; now they happen twice a week. Two decades ago, fewer than 1,000 active satellites orbited Earth. Today, there are more than 16,000 — with SpaceX proposing a million more. Last year, a startup landed a commercial spacecraft on the Moon.

This mad dash for the heavens isn’t being led by governments, the way it was at the dawn of the space age. It’s being led by private companies — run by entrepreneurs and scientists, moving at their own pace, answering to their own boards.

For this series, we’re asking two questions. What are the ways going to space could benefit all of us here on Earth? And — the treaty meant to govern all of it hasn’t been meaningfully updated in nearly sixty years. So who, if anyone, is writing the new rules?

This is our tenth series. You can find previous series and learn more about Solving For at solvingfor.io. You can also listen to this story — narrated by me — by selecting the voiceover at the top of this page.

(Was this forwarded to you? Sign up.)

The British East India Company was one of the most powerful private enterprises ever created.

Founded in 1600 by roughly 200 investors, it received a monopoly from the British crown to trade east of the Cape of Good Hope. Over more than two and a half centuries, the largely unregulated company — run out of a small London office — expanded across the Indian subcontinent and beyond. It built a private army larger than Britain’s standing army, minted its own currency, collected its own taxes, waged wars and governed tens of millions of people. Along the way, it made shareholders fabulously rich, creating fortunes that reshaped Britain’s social order.

Edmund Burke called it “a state in the guise of a merchant.”

Eventually, it became too much: too powerful, too violent, too self-serving, too corrupt. After years of scandal and outrage, the British government stripped the company of its military and political powers in 1858. Parliament formally dissolved the company in 1874.

But before it became an instrument of conquest, the East India Company was something simpler: a platform. It moved goods, capital, and people across distances few could otherwise cross — and from that reach it created forms of commerce that had never existed before.

Subscribe now

A century and a half after the company’s demise, on the day SpaceX went public in June — in the largest stock offering in history — one of the space company’s longest-serving board members went on CNBC and told viewers what investors were buying.

“Think about this as some combination of owning the British East India Company, as you colonize both the solar system and then beyond, and the Moon mission,” said Antonio Gracias, a SpaceX board member and early investor.

Where the East India Company built a platform for trade and settlement east of the Cape of Good Hope, SpaceX is building one for the far reaches above. Instead of ships sailing between London and the Indian Ocean, its reusable rockets make one trip after another to space.

In the first half of 2026 alone, SpaceX launched 78 rockets into space — what was once a momentous event now takes place about every two days. And while SpaceX is dominant, it isn’t alone. Dozens of companies - including Rocket Lab, Planet Labs, Blue Origin and Firefly Aerospace, which put a spacecraft on the moon last year, are racing toward the same goal: making access to space dramatically cheaper, faster and routine.

Yet spaceflight has reached escape velocity while the rules meant to govern it — scant treaties, thin law, limited oversight — remain on the launch pad. One example: more than a million fragments orbit earth, and no single authority is responsible for keeping spacecraft from colliding with them.

The East India Company traded tea, silk, and opium — and ultimately layered war and conquest onto the platform it had built. The question this series asks is what will be built on the new platform reaching toward the heavens. What problems will it solve, what products will it create, and how can it make life better back home on Earth? And, on a frontier with no real rulebook, how can it happen without something going seriously wrong?

For Earth’s Benefit

From the Soviet Union’s launch of the Sputnik satellite in 1957, space belonged almost entirely to governments. It was too costly and risky for any single entrepreneur or company to take on. That began to change on a remote atoll in the Marshall Islands in September 2008, when SpaceX became the first private company to launch a rocket — its Falcon 1 — into orbit.

That day — the fourth, and likely final, attempt by Elon Musk’s SpaceX to put Falcon 1 into orbit — proved to be “an inciting moment,” wrote Ashlee Vance in his book, When The Heavens Went on Sale.

Space shifted from an industry dominated by a handful of slow-moving government operators and contractors to one crowded with private companies eager to launch at an ever faster clip. In the nearly two decades since Falcon 1 reached orbit, dozens of businesses backed by billions of dollars in venture capital, have built rockets, satellites, lunar landers, orbital factories and space stations.

Consider the change in scale: fewer than 1,000 active satellites orbited Earth in 2008; today, there are more than 16,000. More are coming, as China ramps up and private companies multiply. And in January, SpaceX — the market leader by far — filed plans with the U.S. government for its most ambitious project yet: a constellation of one million solar-powered, orbital AI data centers.

Subscribe now

From the beginning, in science fiction and in real life, people have had many reasons to go to space.

There is the pull of discovery epitomized by Gene Roddenbery’s Star Trek and its mission to “boldly go where no man has gone before.” There is national security and national supremacy. In May 1961, less than two months after Soviet cosmonaut Yuri Gagarin became the first human to orbit Earth, President John F. Kennedy — fearful that the U.S. was falling behind — called on the country to put a man on the moon “before this decade is out.” And there is the ambition to make humanity multi-planetary, a vision that runs from Isaac Asimov’s science fiction writings to Elon Musk’s plans today.

But for many companies in today’s space race — and the focus of this series — the animating force is more immediate: The best way to improve life here on Earth is to build something out there.

Rocket Lab, which reached orbit in 2018 and launched many missions since, expresses its vision in ten words: “We open access to space to improve life on Earth.” Planet Labs, operator of more than 200 satellites, says its mission is to “use space to help life on Earth.” Jeff Bezos’ Blue Origin, whose New Glenn rocket first reached orbit in 2025, uses the tag line: “For the benefit of Earth.”

Bezos himself has described a future in which mining, large-scale manufacturing, chemical production take place in space, with finished goods shipped back to earth.

“We need to take all heavy industry, all polluting industry, and move it into space,” Bezos said. “And keep Earth as this beautiful gem of a planet that it is.”

Problems to Solve

Space is already embedded in daily life. Satellites feed the forecasts that farmers plant crops around and insurers use to price policies. Shipping companies track vessels and cargo from orbit. Disaster agencies watch wildfires and hurricanes unfold in real time. GPS tracks Uber rides.

The modern space race could expand that role dramatically — in some cases, it already has.

The Digital Divide

Roughly a quarter of the world’s population still lacks reliable internet access. The gap is especially wide in rural and hard to reach places, where the economics of laying fiber and building cell towers often don’t work. That may be about to change.

Starlink, SpaceX’s broadband satellite network, first started putting satellites into space in 2019 and now has more than 10,200 circling the globe. It has passed 12 million subscribers and operates in 167 countries. Competitors are coming: Amazon is building its own broadband constellation, called Amazon Leo, though it remains far behind Starlink. AST SpaceMobile, meanwhile, is building a satellite network that connects directly to cell phones, with no dish or terminal required. It now has 13 satellites in orbit with a target of more than 40 by end of the 2026.

Space may be where that thirty-year connectivity gap finally closes.

Seeing the Earth

The same orbital vantage point that feeds weather forecasts lets us do something else: see the world as it is.

Planet Labs — now known simply as Planet — was founded in 2010 to put low-cost satellites, called Doves, into orbit and photograph every part of the Earth, every day. The theory was that seeing change as it happened could prompt people to act. The company declared its satellites could “monitor deforestation in Africa, track illegal fishing, and monitor ice caps melting.”

That kind of view had largely been limited to governments and intelligence organizations — and even they lacked a continuous, daily picture of the entire planet. Planet Labs made it widely available. The company now operates more than 200 satellites and images the whole Earth every day, influencing work from oil trading to human-rights advocacy.

The impact became especially vivid in February 2022. As Russian forces massed on Ukraine’s border, Planet’s satellites captured the buildup before the invasion and then documented the war day by day. The images helped investigative outlets like Bellingcat track troop movements, possible war crimes and smuggled shipments using a tool once reserved for spy agencies.

Manufacturing

Some materials crystallize differently in microgravity, forming structures that cannot be easily replicated on Earth. Why not manufacture pharmaceuticals in space?

That is what Varda Space Industries, founded in 2021, is trying to do. It has already demonstrated the basic process: launching a capsule aboard a SpaceX rocket, having it released into orbit, growing pharmaceutical crystals in microgravity, and returning them intact to Earth.

In May, Varda struck its first deal with a major pharmaceutical company, United Therapeutics, to study whether microgravity can improve a drug formulation for a rare lung disease. The first flight is scheduled for early 2027. Nothing has been approved, and no patient has taken a medicine manufactured this way. But for the first time, a pharmaceutical company is spending its own money to find out whether the idea works.

Data Centers

AI data centers are straining power grids and water supplies so severely that more than a dozen states are weighing restrictions on their construction. A solution: data centers in space.

In November 2025 Google unveiled Project Suncatcher, an effort to place satellites carrying the company’s AI processors in orbit, where they could draw on abundant solar power. A prototype may launch as early as 2027.

SpaceX is moving on an even faster timeline and at far greater scale. Up to a million of the space-based AI data centers, called Starmind, are planned. On SpaceX’s first investor call — on August 4 — since going public, Musk said he expects to “start launching these next year.”

Warning Signs

In the digital era, the speed of change has repeatedly outpaced government’s ability to keep up. Space is no exception. The rules governing the new platform being built were written largely for a different era. The 1967 Outer Space Treaty remains the cornerstone of international space law. It declares that the exploration and use of space should benefit all countries and bars nations from placing nuclear weapons or other weapons of mass destruction in space. It has been ratified by 118 countries, including every major country operating in space.

But the treaty was built for a two-superpower contest between governments, not a field crowded with commercial operators pursuing orbital resources, the Moon and beyond.

The last major space treaty, the Moon Agreement, was adopted in 1979. Only a small number of countries have ratified it. Since then, countries have relied largely on national laws and nonbinding international guidelines.

Nowhere is the gap more vivid than in space traffic. There is still no global mechanism for coordinating which spacecraft may be about to collide — nothing like an air-traffic-control system for orbit. Instead, warnings and responses can be entirely ad hoc.

Rosanna Hoffman, who leads space law and sustainability at the U.N. Office for Outer Space Affairs, described a case on the podcast Freakonomics Radio that shows just how thin the safety net can be. Returning from meetings with space officials in Ghana in June 2025, she was in the Brussels airport and her cell phone rang. On the other end was a frantic space official from Malaysia, who said one of the country’s few satellites was on a collision course with a North Korean satellite. But Malaysia had no diplomatic relations with North Korea — and therefore no diplomatic channel to its government. U.N. staff scrambled to reach North Korean officials.

“Luckily, really in the 11th hour,” Hoffman recalled, “they were able to move the satellite, and collision was avoided.”

Collisions have happened. In February 2009, Iridium 33, an active U.S. commercial satellite, accidentally struck Cosmos 2251, a defunct Russian satellite, over Siberia. It was the first collision between two intact satellites. The impact produced some 2,300 trackable fragments across low Earth orbit.

Then, in November 2021, Russia deliberately destroyed one of its own defunct satellites in a missile test. The blast created 1,500 pieces of trackable debris and forced astronauts aboard the International Space Station to climb into their docked spacecraft, ready for a quick departure if needed, as the debris cloud passed nearby.

Such events raise the specter of the Kessler syndrome. NASA scientist Donald Kessler described how one collision could lead to another and then another, each scattering debris that causes the next, until an entire band of orbit becomes unusable for generations. The possibility is no longer dismissed as merely theoretical. Starlink satellites alone performed more than 355,000 collision-avoidance maneuvers in the past year — more than three times their 2024 total.

The debris problem does not remain in orbit. Objects are falling back to earth at a rate of roughly one a week, according to The New York Times. Many burn up completely during reentry, but not all. In 2024, a discarded battery from the International Space Station crashed through the roof of a home in Naples, Florida. A large rocket component struck farmland in Kenya. SpaceX rocket fragments have landed in Poland.

The risk to any one person remains small, but it is rising. One FAA study projected that, if current trends hold, debris from reentering satellites could cause an injury or death roughly once every two years by 2035.

Which brings us back to the British East India Company — the company Antonio Gracias, a longtime SpaceX investor and board member, compared SpaceX to. It began as a platform, moving goods, capital, and people across distances nothing else could cross. In time, it became something else — something Parliament ultimately dismantled by force of law.

The modern space race could transform life on Earth. The question is whether we can realize that promise without allowing the platform — and the powers building it — to outrun the rules meant to govern them.

Next up: How we got here — how launching a rocket, once a national feat, became almost routine. None of it happened by accident.

Prefer to listen? I narrate each edition myself. Find the audio at the top of this page or under the Listen tab at solvingfor.io.

Solving For takes on one pressing problem at a time: the stakes, the forces driving it, and what a path forward might look like. Each series unfolds in weekly installments.

Previous series have examined China’s rare earth dominance, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, social media and teen mental health, a world rearming as the global rules-based order weakens, America’s national debt crisis and — most recently - renewing the American Dream. Each series is available for reading or listening (and I narrate them all) at solvingfor.io.

This week we return to The Open Thread, our bridge between series — a space to update past stories and experiment. For this installment we turn back to AI Safety, a subject we first tackled last fall with our three-part series, “The Control Problem: Solving For AI Safety.” Last month we shared a conversation I had at Faena Rose with Tristan Harris, a leading voice on the dangers presented by the way AI is being deployed. And this week, the news handed us a real-world case — seven days that captured the full scope of the challenge.

If you’re new here, Solving For is a newsletter that aims to understand the many challenges we face by going deep on one at a time — the stakes, the forces behind it, and the credible paths forward. Our last series tackled the American Dream. The next one: Space.

If you prefer to listen, find the audio narrated by me at the top of this page. Every Solving For series is available to read or listen to — I narrate each one myself — at solvingfor.io.

In one week this month, the high stakes and steep challenges of building AI safely unfolded like a three-act play. First, the warning. Then the complication. Finally, confirmation.

Act I

On July 14, Demis Hassabis — Google DeepMind CEO and co-founder, Nobel laureate, and AI pioneer — delivered the warning, publishing an essay titled “A Framework for Frontier AI and the Dawning of a New Age.”

Hassabis wrote that humanity is “probably only a few short years away” from Artificial General Intelligence, when AI models will match the human brain’s cognitive abilities.

The magnitude, he declared, will be unprecedented. He put the scale at ten times the Industrial Revolution, arriving ten times as fast. He compared it to the discovery of electricity and the harnessing of fire. The possibilities he laid out: accelerated drug discovery, new clean energy sources, novel materials.

But, he warned, as we get closer, the dangers become greater. Much greater.

Subscribe now

The two primary concerns: misuse and loss of control.

The misuse of AI by people to hack bank accounts at scale, or design bioweapons. The loss of control when AI becomes so powerful it acts on its own, learns on its own (so-called “recursive self-improvement”), and even refuses direction.

Urgent action is needed, Hassabis wrote, to address risks that might arise as we get closer to AGI. Time and space are needed to get this next crucial step right, he declared, before warning: “Currently, as a field and as a wider society, we aren’t doing that.”

These are not the words of an AI doomer — but the CEO of one of the world's leading AI labs.

To get this consequential step right, Hassabis proposed the creation of a Standards Body, led by the US, to develop assessment protocols and review new models before they are released. This, in turn, could lead to the creation of a shared global framework for AI models around the world.

We must use this narrow window, Hassabis urged, before AGI arrives.

Act II

Two days later, on July 16, came the complication. Finding common ground ran up against the reality of this century’s fiercest global rivalry.

Beijing startup Moonshot AI released Kimi K3, an AI model that by one measure — front-end coding — rated number one in the world, more capable than either Anthropic’s Claude Fable 5 or OpenAI’s highest-performing model, GPT-5.6 Sol, which ranked second and third.

The implications go far beyond a benchmark win. The reason is that Kimi K3 is open-weight, meaning Moonshot AI plans to give it away for free — anyone with sufficient computing power can download and run it at no cost. That’s not the case with the top-flight models of Claude, ChatGPT, or Google DeepMind’s Gemini. That difference cuts to the heart of U.S. frontier labs’ business model: why would a company pay a subscription for AI coding if another company — in this case, a Chinese competitor — offers a comparable, even superior, product for free?

The answer is that many U.S. companies aren’t. One example: San Francisco-based food delivery company DoorDash uses an Anthropic model for its cutting-edge AI tasks but Moonshot AI’s less costly models for other work, according to The Wall Street Journal.

This has raised the question of whether U.S. companies should be allowed to use Chinese AI models at all. The U.S. has, for instance, forbidden Chinese car manufacturers, like BYD, from selling cars in the states. But no such restriction exists for Chinese AI models like DeepSeek or Z.ai — anyone, anywhere, can use them, though not without limits: ask Moonshot AI’s Kimi K3 about Tiananmen Square in 1989, and it won’t answer.

China’s AI advances have come despite U.S. export controls restricting the sale of advanced chips to China. But U.S. frontier AI companies have presented another theory for China closing the gap so fast: “distillation.” This is when a less capable AI is built by using outputs from an existing, stronger model. Anthropic charged that China’s Alibaba had attempted what it called “the largest known distillation” attack it’s ever encountered.

U.S. Treasury Secretary Scott Bessent declared sanctions against China “will be on the table” in response to large-scale distillation attacks. Bipartisan legislation is percolating in Congress.

Nvidia CEO Jensen Huang, meanwhile, argued this week that U.S. companies should not be forbidden from using Chinese AI models, and that they don’t present a threat to U.S. frontier labs. “There’s no scenario where China runs U.S. companies off the road,” he told Axios.

But the point is this: amid the worries about who’s ahead, who’s distilling, whether U.S. companies should even be using Chinese models, one question got lost. It was the one Hassabis had raised two days earlier: not about who wins the race, but whether anyone is safely steering it.

Act III

Then five days later, on July 21, came the confirmation.

OpenAI disclosed that two of its own AI models, which it was testing internally, had escaped and hacked into the servers of Hugging Face, a company that hosts AI models and datasets, in order to get the answers to the test they were being given.

If the two AI models were people, it would be like two students breaking into a teacher’s locked desk to get the answers to a quiz.

How it happened is that OpenAI was testing the models in a “sandbox,” an isolated area walled off from the internet. Or so they thought. The AI models breached the wall and — on their own — figured out a way to connect to the internet. From there, the two AI models navigated to Hugging Face, presumably thinking that’s where the answers to the test could be found, and broke into Hugging Face’s computer systems, carrying out more than 17,000 actions.

Hugging Face caught the intrusion first, before discovering that they were OpenAI models. OpenAI released a statement calling it an “unprecedented cyber incident, involving state-of-the-art cyber capabilities.”

Share

What makes it so extraordinary is the absence of a human trying to do bad things. OpenAI had deliberately lowered the models' safety guardrails to stress-test them, but no person directed the models to attack Hugging Face specifically. They were simply pursuing a specific goal — to pass the test — with such focus and determination that they found a way onto the open internet and hacked into Hugging Face. It was a means to an end.

With misuse and loss of control representing two primary concerns, the single incident represented both. Humans lost control of two AI models, which acted on their own. Then misuse was the result.

Ultimately, no real damage was done. The OpenAI models were trying to pass a test. But the capability it revealed previews a result that could genuinely be destructive. AI models that find their own way past walls that human experts thought were secure, in pursuit of a goal as trivial as passing a test, raise an obvious and uncomfortable question: what happens when the goal isn’t trivial, and no one’s watching as closely?

Epilogue

Which brings us back to Hassabis, just a week before the zero-day breach was revealed. He closed his essay by saying the future isn’t fixed — that AGI could unlock a new era of scientific discovery and human flourishing, if we get this transition right. Nothing about these seven days suggests anyone’s found the answer yet.

In our previous three-part series on this question, The Control Problem: Solving For AI Safety, the solution we hit upon was found in nuclear non-proliferation and 1968. For some two decades, Washington and Moscow treated nuclear weapons as the other side’s problem — as a race to be won, a threat to be outpaced, not a danger they shared. The Nuclear Non-Proliferation Treaty, signed in 1968, only became possible once that framing broke.

AI hasn’t made that shift yet. Hassabis’s Standards Body proposal is a step toward a 1968 moment. This time, though, it would be the U.S. and China that would need to sketch out shared rules governing AI’s deployment and growth. Hassabis said the window to do that is closing as the arrival of AGI approaches. The Hugging Face breach makes the point plain: this isn't one country's problem to solve. It's everyone’s.

Whether AI makes the turn that nuclear weapons eventually made — from rivalry to shared restraint to global accord — is still the question, and this week’s answer was rivalry. The window that Hassabis described is still open.

Prefer to listen? I narrate each edition myself. Find the audio at the top of this page or under the Listen tab at solvingfor.io.

Solving For explores credible solutions to the challenges we face — one pressing problem at a time, through deeply reported series that trace the stakes, the forces behind it, and ways forward.

Previous series have examined China’s rare earth dominance, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, social media and teen mental health, a world rearming as the global rules-based order weakens, America’s national debt crisis, and — most recently — how to renew the American Dream for a new era. Each series is available for reading or listening (and I narrate them all) at solvingfor.io.

This week, we conclude our series on the American Dream.

(If you’re new here, Solving For is a newsletter exploring one pressing problem at a time — the stakes, the forces driving it, and what credible solutions might look like.)

In part one, we unpacked the problem: Upward mobility has weakened dramatically over the past half-century, and it’s weighing heavily on Americans. A poll last year found that nearly 7 in 10 believe the American Dream no longer holds true — or never did.

In part two, we examined how we got here: automation and trade eliminating middle-income jobs that once served as rungs on the ladder to the middle class; the soaring cost of college and the student debt that followed; tax policy favoring investment over work; and declining trust in the very institutions needed to build things back.

Now, as America searches for the shape of its next political era, it must find a way to make broad upward mobility—the heart of the American Dream—its defining purpose.

So this week, we turn to solutions: five pillars for renewing the American Dream in a new era, as artificial intelligence promises extraordinary abundance while threatening to widen the divides already holding people back.

You can listen to this Solving For story—narrated by me—by selecting the voiceover at the top of this page. You can also visit solvingfor.io, where all narrated stories and previous series are organized under the Listen tab.

(Was this forwarded to you? Sign up.)

“I was born, I have lived, and I have been made over.”

That is how Mary Antin began The Promised Land, her 1912 memoir of coming to America. Born into a Jewish family in the Russian Empire, in what is now Belarus, she arrived in Boston at thirteen speaking little English. Public schools and the Boston Public Library changed the course of her life.

By the end of her memoir, Antin believed America had given her something greater than economic opportunity: the freedom to imagine a future unconstrained by her past.

She put it in a single sentence: “Mine is the whole majestic past, and mine is the shining future.”

Nearly twenty years later, historian James Truslow Adams quoted that line in The Epic of America, the Depression-era book that introduced the phrase “the American Dream.”

To Adams, Antin’s declaration captured the nation’s defining promise: not that success was guaranteed or measured by vast wealth, but that in America, your future did not have to be determined by your past.

It’s a measurable promise. Roughly nine in ten children born in 1940 grew up to earn more than their parents. Today, that figure is about 50 percent — little better than a coin flip. From 9 in 10, to 1 in 2.

As America begins its next 250 years, the challenge is becoming more urgent.

In his 2026 letter to investors, BlackRock CEO Larry Fink named three forces shaping the economy. One was shifting global trade. The other two strike directly at the American Dream: “the inequality that’s risen over the past generation,” and “how AI threatens to widen the gap.”

The question, then, is not how to restore the American Dream. It’s how to renew it for a new era.

Toward a New Political Order

America’s enduring strength has always been twofold: innovating and growing, while spreading opportunity so each generation can climb higher than the last. It continues to excel at the former, but struggles at the latter.

That rupture has reshaped American politics. We’re now in what Ezra Klein and Derek Thompson call in their book, Abundance, “a messy interregnum between political orders.”

Historian Gary Gerstle defines a political order as a governing consensus both parties largely accept. The New Deal order, born in the 1930s, held that capitalism required government oversight and basic economic security. Franklin Roosevelt built it, but Republicans largely accepted its foundations. Eisenhower preserved Social Security. Nixon proposed versions of guaranteed income and universal health insurance.

Beginning in the late 1970s, that consensus gave way to the neoliberal order: markets allocate resources better than government, regulation lighter, trade and globalization expanded, taxes lowered. Ronald Reagan became its defining figure, but Bill Clinton entrenched the order, embracing free trade, welfare reform, financial deregulation, and declaring “the era of big government is over.”

That order fractured in the 2010s in the wake of Iraq War, 2008 financial crisis, globalization, and deepening inequality. Too many communities and workers felt left behind. The American Dream — that each generation should do better than the last — became less certain, fueling populist revolts across the political spectrum.

Today we’re living through the collapse of one political order before another has fully emerged.

If the American Dream is to endure, the next political order must once again make broad upward mobility—not simply economic growth—its defining purpose. That search doesn’t divide neatly along partisan lines. It points toward five enduring pillars: ownership; building, not blocking; connections to opportunity; fairness; and trust.

None of these ideas is new. Some call for more government, others for less. What unites them is not ideology, but evidence.

Each is, in its own way, an opportunity—as Mary Antin wrote—to be “made over.”

Ownership

If the next American Dream is about broadening opportunity, it must also broaden ownership. One way to do that: give every American child a stake in the country’s future.

On July 4, the federal government began seeding an investment account for every newborn. Invest America accounts — so-called “Trump Accounts” — provide a one-time $1,000 contribution invested in a stock index, with families, employers, and others able to contribute up to $5,000 a year. Philanthropists and businesses have pledged more, including Michael and Susan Dell, who committed $6.25 billion for children in lower-income communities.

The premise is simple: assets shape lives differently than income. A nest egg can help pay for college, buy a first home, start a business, fund retirement. It also gives young people something they know is theirs and can build on.

The idea stretches back centuries. In 1797, Thomas Paine proposed a birthright endowment for every young adult, funded by inherited wealth. Nearly two hundred years later, sociologist Michael Sherraden argued in his 1991 book Assets and the Poor that owning assets changes how people think about the future. In 2010 economists Darrick Hamilton and William Darity Jr. developed the modern idea of “baby bonds” as a way to narrow wealth inequality. In 2023 Connecticut became the first state to try it, seeding $3,200 for each Medicaid-covered baby.

The strongest evidence comes from Oklahoma. In 2007, researchers randomly assigned newborns to either receive a seeded college savings account, or no account. The study, SEED for Oklahoma Kids, was designed by Sherraden’s Center for Social Development at Washington University in St. Louis. It found that children with investment accounts didn’t just accumulate more savings — they also showed greater hope, stronger educational expectations, and fewer behavioral problems, while their parents were more likely to prepare for college.

The evidence suggests these programs work best when they are universal, automatic, and progressive. Invest America accounts are universal and moving toward automatic enrollment. What’s missing is progressivity. Every child receives the same $1,000 regardless of family wealth, allowing those with the greatest means to build much larger balances over time.

To endure over the long term, the program will likely need a name that isn’t polarizing. But, in many ways, the idea is a twenty-first century complement to Social Security. Roosevelt’s landmark program, signed into law in 1935, provided income at the end of life. Baby bonds seek to provide capital at the beginning.

Building, Not Blocking

Opportunity expands when it’s easier to build. It generates more jobs and investment, while greater supply can lower prices.

In June 2025, California Gov. Gavin Newsom signed sweeping legislation exempting many housing projects from the state’s landmark environmental review law while streamlining approval of infrastructure projects, calling it part of “a budget that builds.” He credited the shift to Abundance, the 2025 book by Klein and Thompson, which argues that decades of well-intentioned regulation have made it unnecessarily difficult to build the homes, energy projects, and infrastructure people need.

The result: fewer homes, higher prices, and diminished opportunity.

Since the 1970s, local land-use rules have made housing slower and more expensive to build, contributing to a shortage of roughly six million homes, according to the American Enterprise Institute. The conservative AEI’s prescription—smaller lots, by-right approval for duplexes and townhomes, streamlined permitting—is similar to the Abundance movement’s, despite arriving from the opposite side of the political spectrum.

A similar principle applies to people’s ability to move into new work.

Occupational licensing serves an important purpose: protecting the public by ensuring professionals meet minimum standards. But licensing has expanded well beyond health and safety. In some states an aspiring manicurist must complete hundreds of hours of training before legally earning a paycheck. Today, occupational licensing covers nearly one-quarter of the U.S. workforce, up from about 5 percent in the 1950s.

Economists Morris Kleiner and Ming Xu found that workers in licensed occupations are 24 percent less likely to switch careers than comparable workers, suggesting licensing can become a barrier to new opportunities. The Obama administration reached much the same conclusion in a 2015 review, finding that excessive licensing requirements can raise prices and restrict employment opportunities.

Taken together, these examples suggest an emerging consensus across the political spectrum: expanding opportunity often begins by reducing unnecessary barriers — to building homes and infrastructure, and to building a career.

Connections to Opportunity

Mobility depends on more than economic growth. It requires connecting people to opportunity.

Economist Raj Chetty, who directs Harvard’s Opportunity Insights lab, has shown that upward mobility varies enormously from one neighborhood to the next, even within the same city. Much of that difference is social isolation: children, growing up just a few miles apart, have access to very different networks and opportunities.

But the right interventions, Chetty’s research shows, can break that isolation.

A recent example comes from HOPE VI, the federal program that spent $17 billion replacing 262 distressed public housing projects across the U.S. with mixed-income developments. In a January 2026 paper, Opportunity Insights found that the redevelopment itself did little for adults who already lived there. But their children did dramatically better: kids who spent their entire childhood in a revitalized development earned roughly 50 percent more as adults than children who stayed in unreformed public housing.

The mechanism wasn’t the buildings, it was who lived near them. Before revitalization, researchers described these sites as islands of disadvantage, where kids in public housing interacted almost exclusively with other kids in public housing. HOPE VI drew higher-income residents into the same neighborhood, and that proximity broke the isolation. Children began forming friendships across income lines, and those cross-class connections are what the data ties to their later success.

“Place matters for mobility,” Chetty said. “But to a large extent this is because some places are more connected than others.”

Fairness

A thriving economy should reward work as much as wealth. Yet income from work is taxed more heavily than investment.

A worker’s wages are taxed at a top federal rate of 37 percent. Long-term capital gains—the profits from selling stocks, real estate, or a business—are taxed at 20 percent. That gap has become much more consequential as more fortunes are built by stock holdings than paychecks.

Journalist David Leonhardt, in his book Ours Was the Shining Future, argues that this is one of the defining fault lines in modern American capitalism. Over the past half-century, the economy has increasingly rewarded ownership over work. The tax code is hardly the only reason, but it is one of the clearest expressions of that choice.

Vinod Khosla — one of Silicon Valley’s most successful investors, who recently agreed to purchase the NFL’s Seattle Seahawks — argued in a recent interview with The Wall Street Journal that preferential tax treatment for capital made sense when investment was scarce. That world is disappearing. And it’s particularly the case as AI increases the returns to capital while reducing the relative value of labor, he argues. As a result, he said, capital gains and ordinary income should be “taxed the same.”

“We need to tilt the seesaw toward labor,” he added.

OpenAI made a similar case in an April white paper, Industrial Policy for the Intelligence Age: Ideas to Keep People First, warning that AI could expand corporate profits and capital gains while shrinking wage income and the payroll taxes that fund programs such as Social Security and Medicare.

But the deeper issue isn’t just economic. It’s about fairness too.

Every political order ultimately depends on people believing the rules are fair. So does the American Dream. Upward mobility is built not only on opportunity, but on the conviction that hard work still gives ordinary people a fair shot at getting ahead.

Trust

Trust is what makes the four pillars possible. Unlike tax rates or investment accounts, trust cannot be created by legislation. It is earned slowly—through governments that keep their promises, institutions that work, and communities where people experience competence and fairness in everyday life.

One way to catalyze it is through an idea championed by the late Senator John McCain. In 2020, two years after his death, the McCain-Reed Commission published the Inspired to Serve report, which found that the desire to serve — whether military, national, or public service — outpaces the opportunity to do so. Its central recommendation: expand access so 5 million Americans begin some form of service each year by 2031, including a nearly tenfold increase in national service slots.

The case isn't sentimental. Chetty's research shows that economic connectedness — friendships that cross class lines — drives upward mobility. National service creates exactly that kind of contact by design: people who would otherwise never meet, working side by side. Over time, it builds trust itself, reinforcing that we’re all part of the American enterprise.

Larry Fink’s warning is worth returning to. Artificial intelligence promises extraordinary abundance. It also threatens to concentrate wealth and opportunity on an even greater scale. Whether AI expands the American Dream or deepens its decline will depend less on the technology itself than on the choices we make about the society that surrounds it.

Every idea here — broadening ownership, making it easier to build, creating more pathways to opportunity, ensuring fair rules, building trust — ultimately depends on people believing those systems can work for them. Without that confidence, even the best reforms will struggle to take root.

Mary Antin believed “the whole majestic past” belonged to her, and so did “the shining future.” James Truslow Adams recognized she was describing something larger than one immigrant’s experience. She was describing America’s promise.

The American Dream isn’t something we preserve. It’s something we build, together, every generation.

Put another way: Ours is the whole majestic past. And ours can still be the shining future.

Prefer to listen? I narrate each edition myself. Find the audio at the top of this page or under the Listen tab at solvingfor.io.

Solving For takes on one pressing problem at a time: what’s broken, what’s driving it, and what a path forward might look like. Each series unfolds in weekly installments.

Previous series have examined China’s rare earth dominance, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, social media and teen mental health, a world rearming as the global rules-based order weakens, and — most recently — America’s national debt crisis. Each series is available for reading or listening (and I narrate them all) at solvingfor.io.

This week, we continue our series on the American Dream. It feels like the right conversation to have as America marks its 250th birthday — a moment to celebrate our country's extraordinary achievements and reflect on one of its defining ideals. (If you’re new here, Solving For is a weekly newsletter examining one pressing problem at a time — what it is, why it happened, and what credible solutions might look like.)

Last week, we looked at the problem. The data showed that upward mobility has weakened dramatically over the past half century.

This week, we turn to how we got here — the context and forces at work. Next week, we'll explore ways forward.

You can read or listen to every Solving For series — narrated by me — at solvingfor.io. Click the article voiceover at the top of this page to listen to this story.

Learn more, and find previous series. (Was this forwarded to you? Sign up.)

David Leonhardt had a question.

The New York Times writer wanted to know: did American children still end up better off than their parents?

He put it to Raj Chetty, then a young economist. Chetty was one of a handful of researchers given access to a massive database of anonymized IRS records that had long been off-limits. But the records only went back to the early 1990s. Nowhere near enough to compare one generation against the next.

Chetty said he and his colleagues would see what they could do.

Their answer: pair decades-old census records with the newer tax data. Track what a child’s household earned while they were growing up. Then find that same child’s income as an adult. Adjust for inflation. Repeat it for millions of families.

It was detective work in the social sciences. He had two sets of records nobody had cross-referenced at this scale, and a hunch that combining them could reveal something new.

The method worked.

What it uncovered laid bare one of the defining challenges facing America today.

Subscribe now

For children born in 1940, roughly nine in 10 grew up to earn more than their parents. For children born in 1984, only about half did.

Put differently, what was once a near certainty had become little better than a coin flip.

The findings were published in Science in 2017. Leonhardt would later tell the story in his 2023 book, “Ours Was the Shining Future.”

The numbers are worth sitting with, especially this week.

America was founded on the radical idea that opportunity — not birth — should determine a person’s future. Yet as the nation celebrates its 250th birthday, the evidence is that birth circumstances have become more predictive of where Americans end up.

Progress has been America’s “lingua franca,” Leonhardt wrote in his book. The U.S. origin story is about throwing off a great power, then eclipsing every other one that came after. Defeating fascism and communism. Putting a man on the moon. Creating the internet. Perhaps most impressive of all, building a massive middle class.

That spirit hasn’t disappeared. America remains the world’s leading engine of entrepreneurship, attracting 64 percent of global startup funding in 2025, up from 56 percent the year before. The country continues to invent, build, and generate enormous wealth. What it no longer does as reliably is spread opportunity broadly enough to ensure that the next generation does better than the last.

Chetty's longitudinal study quantified that shift. It showed that one of America’s defining promises had been interrupted.

The question was why?

Rungs on the Ladder Disappeared

The American Dream has not disappeared. We all know people who represent it, or are on their way to achieving it. But over decades, the Dream has been weakened by powerful forces. Five stand out.

After World War II, the American economy produced something unusual in human history: millions of workers without college degrees could still climb into the middle class. Manufacturing, construction, clerical work and skilled trades provided stable, middle-income jobs that allowed Americans to buy a home, raise a family, and expect their children to do even better. They were the rungs on the ladder.

Starting in the 1980s and accelerating through the 2000s, many of those rungs disappeared.

Two forces removed them. The first was technology. Computers and automation didn’t simply replace workers — they replaced specific kinds of work. Routine tasks, whether performed on assembly lines or in back offices, were the most vulnerable. What grew instead were jobs at the top and bottom of the labor market: high-skill work requiring education, and low-paying service work that could not easily be automated. The middle hollowed out. Labor economists call this phenomenon job polarization: the rungs didn’t just get harder to climb — they stopped existing.

The second force was trade. In 2001, China joined the World Trade Organization. Economists David Autor, David Dorn, and Gordon Hanson tracked what happened in communities most exposed to Chinese manufacturing competition. Local labor markets adjusted remarkably slowly. Wages stayed depressed and employment stayed low at least a decade after the shock arrived, or never recovered.

Most research agrees automation did more national damage than trade. A Ball State University analysis attributes 85 percent of manufacturing job losses to productivity gains, not trade. But trade had outsized, life-altering impacts in specific towns and regions, shuttering plants and wiping out jobs. Autor, Dorn, and Hanson found that Chinese import competition, while responsible for a smaller slice of manufacturing losses nationally — roughly a quarter, by their count — hit specific communities all at once, and many of those places are still recovering.

For decades, America had relied on millions of middle-skill jobs to turn economic growth into upward mobility. As those jobs disappeared — whether through automation, trade, or both — so did one of the country’s most reliable pathways to the American Dream.

The Credential Bottleneck

When the rungs disappeared, something had to replace them. College filled the gap, becoming the default pathway to the middle class — the most reliable route to upward mobility.

The problem was that the ticket got more expensive at the same time everyone decided they needed one — and the most valuable seats barely expanded to meet the demand.

In 1980, attending a four-year college cost roughly $10,000 a year, including tuition, fees, room and board, adjusted for inflation. By the 2025–26 academic year, that same combination — tuition, fees, room and board — had risen to nearly $26,000 at in-state public four-year institutions, and to about $61,000 at the average private nonprofit college. At the most selective schools, such as the Ivy League, the same core costs now run $83,000 to $94,000, according to each institution's own published pricing. Even using the public in-state figure, the price of the credential had risen roughly two and a half times in real terms.

At the same time, many elite institutions chose not to expand enrollment, preserving the scarcity that reinforced their prestige. Harvard's freshman class has stood at roughly 1,600 students for more than four decades — while applications grew substantially over the same period. Before U.S. News began ranking colleges in 1983, schools like Harvard, Yale, and Stanford were expanding enrollment. The rankings changed the incentive: selectivity became a part of the product. Low acceptance rates drove prestige, and prestige drove applications, donations, and pricing power. Scarcity became an asset. Between 2002 and 2019, Ivy League acceptance rates were cut roughly in half.

In response, families spent more to gain an edge — tutors, test prep, curated extracurriculars — widening the advantage of those who could afford the race. The so-called “Varsity Blues” federal prosecutions in 2019 illustrated the lengths — and lies — wealthy parents and admission consultants would go.

Meanwhile, colleges across the board continued raising tuition, pushing students to borrow ever larger sums. By 2010, student loan debt in the United States had surpassed credit card debt for the first time.

The debt burden has grown so large that millions of Americans now carry student loans well into middle age, with growing numbers still repaying them into retirement.

At precisely the moment America needed more rungs on the ladder, its most prestigious universities chose to preserve the scarcity central to their prestige rather than expand access to meet the need. One of the country's most powerful pathways to upward mobility became narrower just as it became more essential.

Geography Became Destiny

The collapse didn’t happen everywhere at once. It happened in places — in specific towns and regions, along the rivers and rail lines where the mills and factories had once stood.

As manufacturing declined, the new economy concentrated elsewhere. Technology, finance, health care, and professional services clustered in a relatively small number of metropolitan areas — San Francisco, New York, Boston, Seattle, Austin, among them. Workers with the education and skills those industries demanded followed the jobs. Those without them often stayed behind.

The dynamic became self-reinforcing. Places with talent and capital drew more of both — economists call it agglomeration. Places that lost their anchor employers lost the tax base and demand that might have replaced them. One rose. The other declined. The pattern earned its own name: “winner-take-most.”

America didn't just grow more unequal. It grew unequal by ZIP code.

Chetty’s Opportunity Atlas — built from the anonymized tax and Census records covering nearly the entire U.S. population — showed just how much place itself matters. Researchers traced children’s economic outcomes back to the neighborhoods where they grew up. Children from families with similar incomes, living only a few miles apart, often had dramatically different chances of reaching the middle class or beyond. Where you grew up increasingly shaped where you ended up.

Communities that lost employers often lost tax revenue, social capital, and opportunity alongside them. Chetty’s research found that neighborhoods with strong schools, high employment, and robust social networks consistently produced better outcomes for children. Upward mobility had become deeply tied to geography.

Policy Locked It In

None of these changes made the American Dream’s decline inevitable. Labor markets change. Institutions adapt. Governments intervene. Communities reinvent themselves. What happened instead was that the rules of the economy were rewritten — gradually, across both parties and several decades, to increasingly favor those who already had the greatest advantages.

The clearest example is the tax code. The top federal income tax rate stood at 70 percent in 1979. By 2007, the top rate had been cut in half, to 35 percent. The income share of the top 1 percent more than doubled, from about 10 percent in 1979 to more than 20 percent in 2007, according to the Congressional Budget Office.

Money made from investments has long been taxed more lightly than money made from a paycheck — that preference dates to the 1920s. But as the wealthy increasingly made their money from investments and business ownership instead of salaries, that old preference came to shelter a much bigger share of their income. Today the top rate on a paycheck is 37 percent. The top rate on long-term investment gains is 20 percent — roughly half.

Workers lost bargaining power just as capital gained it. Union membership fell by more than half in the four decades after 1980 — from 22.2 percent of workers in 1980 to 9.4 percent by 2022. The gains increasingly flowed to owners and shareholders instead.

Then came the financial crisis of 2008.

Workers entering adulthood inherited an economy that had already lost its traditional pathways upward. The Great Recession narrowed them further. Millions graduated into the weakest labor market since the Great Depression.

Why None of It Got Fixed

Each of the forces above was visible. Documented. Named. The study by Chetty and his colleagues is nearly a decade old. It hasn’t been fixed.

A big part of the explanation is trust — or, more precisely, its erosion.

In 1958, 73 percent of Americans said they trusted the federal government to do the right thing most or almost all of the time. By 2025, that number was 17 percent, according to Pew.

Some of that collapse was earned. The Vietnam War and the Watergate scandal gave real reasons for people to stop believing what their government told them.

But the erosion didn’t stop where earned distrust left off. Ronald Reagan made a more explicit push in 1981, declaring: “Government is not the solution to our problem; government is the problem.”

Donald Trump has gone further, actively sowing distrust. He hasn’t just cast the government as inefficient or overreaching — the traditional conservative critique. He has cast the mechanisms of democracy itself as illegitimate, calling elections “rigged” — unless he, or candidates he supported, won.

Together, these forces created a vicious cycle. The economy stopped delivering for many people, giving them reason to distrust the institutions charged with governing it, making it harder to build the coalitions required to fix what broke, giving people still more reason to distrust.

The ladder got shorter. So did trust in the institutions that could rebuild it.

Up Next: Examining the solutions.

Prefer to listen? I narrate each edition myself. Find the audio at the top of this page or under the Listen tab at solvingfor.io.

Solving For takes on one pressing problem at a time: what’s broken, what’s driving it, and what a path forward might look like. Each series unfolds in weekly installments.

Previous series have examined China’s rare earth dominance, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, social media and teen mental health, a world rearming as the global rules-based order weakens, and — most recently — America’s national debt crisis. Each series is available for reading or listening (and I narrate them all) at solvingfor.io.

On Dec. 19, 1966, my wife’s mother, father, and two sisters walked through the doors of the Freedom Tower in downtown Miami. They had fled Cuba and the Castro dictatorship with three changes of clothing — and a dream.

America gave them the chance to realize it. They settled in New York City, became U.S. citizens, worked multiple jobs, and built a good middle-class life. My wife, Danet, was later born in Queens. Their story is one example of an idea that has propelled this country from its earliest days: the American Dream.

Yet that dream is now under pressure unlike any time in decades. As America readies to celebrate its 250th birthday, the data tells a sobering story — about upward mobility stalled, opportunity narrowed, and a promise that fewer and fewer Americans believe still holds.

This series is about that gap. Today, we examine the problem. Next week: the forces driving it. The week after: what solutions might actually look like.

If you prefer to listen, find the audio narrated by me at the top of this page. Every Solving For series is available to read or listen to — I narrate each one myself — at solvingfor.io. (Was this forwarded to you? Sign up here.)

The most invoked phrase in American political life was born not in triumph, but in crisis.

In the spring of 1931, the country was gripped by the Great Depression. Millions were out of work. Unemployment would surge to a quarter of the workforce. Breadlines stretched around city blocks. Banks collapsed at a record clip, prompting bank runs as desperate account holders withdrew deposits.

Into that moment, a historian named James Truslow Adams sat down to finish a book about his country. One that traced the arc of American history and arrived at a conclusion: the nation's greatest contribution to the world was not its wealth or its military power, but an idea.

He called it the American Dream.

Not a dream of material abundance or outsized wealth, but a dream in which every man and every woman could attain the fullest measure of their potential, regardless of the circumstances of their birth. This, he wrote, is “the greatest contribution we have as yet made to the thought and welfare of the world.”

Adams wanted to call his book the American Dream. But his publisher didn't think the phrase would sell. The book was instead called The Epic of America. It was a national bestseller. The title was largely forgotten. The idea inside it was not.

Subscribe now

Yet it’s now invoked so often — in presidential speeches, immigration ceremonies, commencement addresses, mortgage company ads — that few know its origin.

The idea that has fueled this country — that where you start shouldn't determine where you finish — is failing by its own terms. And the data is unambiguous.

A survey last year found that nearly 7 in 10 Americans say the Dream — that if you work hard, you will get ahead — no longer holds true, or never did. That is the highest share in nearly 15 years of Wall Street Journal-NORC polling. Only 25% say they have a good chance of improving their standard of living, a record low in surveys dating to 1987.

The sentiment runs deeper than economics. In a March 2026 NBC News poll, 59% of registered voters agreed that the economic and political systems are stacked against people like them — tying a record high over roughly 40 years since NBC News started polling the question. For context: that’s higher than in the immediate aftermath of the Great Recession.

The pessimism cuts across party lines, generations, and income levels. But it is not evenly distributed. An AP-NORC poll conducted this spring found that nearly half of Americans 60 and older still believe the Dream holds true. Among adults ages 18 to 29, that number falls to one in five.

In 2024, Open Research published findings from a three-year economic study. One participant, a woman identified in the study as Tara, put it plainly: “If I work hard, I’m gonna be successful, I’m going to make it.”

Then she added, “I think it’s a bit delusional actually, now that I said it out loud.”

The Dream Wasn’t for All

The Dream that Adams named was always a specific promise. It was never universally applied.

The country that produced it also produced slavery, which lasted until 1865. Women couldn’t vote until 1920. Legal racial segregation persisted until 1964. The postwar economic boom that many Americans remember as the Dream’s fullest expression — the suburbs, the union jobs, the upward mobility — was built on federal policies that systematically excluded Black Americans from its benefits.

The Dream, in other words, was never as advertised. But the promise embedded in it was real, and worth fighting for. Generations did exactly that.

They include people like Irving Berlin, whose family fled Russia when he was five to escape anti-Jewish pogroms. Speaking no English when he arrived, he grew up to write the songs that became the soundtrack of American optimism, including “God Bless America.” And people like Madam C.J. Walker, born to sharecroppers in Louisiana in 1867, two years after emancipation, who built a hair care business into a fortune that made her — a Black woman from the segregated South — one of the first self-made female millionaires in American history. Their climbs required beating odds that shouldn’t have existed. For every Berlin or Walker, millions more never got the chance.

Subscribe now

Langston Hughes saw it clearly. In his 1951 poem Harlem he asked: What happens to a dream deferred? A society that promises the Dream to everyone and delivers it to some is not keeping its promise — it is deferring it, indefinitely, for the rest.

Martin Luther King Jr. called it a broken contract. At the Lincoln Memorial in 1963, in his “I Have a Dream” speech, he told the crowd that the country’s founders had signed a check — one that guaranteed every American the unalienable rights of life, liberty, and the pursuit of happiness. Black Americans, he said, had been handed a check marked insufficient funds.

In a July 4, 1965 speech on the American Dream, King said that “ever since the Founding Fathers of our nation dreamed this dream, America has been something of a schizophrenic personality.” He added: “Slavery and segregation have been strange paradoxes in a nation founded on the principle that all men are created equal.”

The promise, across generations and across the battles fought to redeem it, was the same: birth shouldn’t determine destination. Not a guarantee of wealth, but a guarantee of access to the ladder, and a fair shot at climbing it.

Yet even as more Americans gained access to the ladder, the climb has gotten harder.

The Math Behind the Dream

The data Raj Chetty, a Harvard professor, and his colleagues at Opportunity Insights have assembled over the last decade is the most comprehensive picture we have of what the Dream actually delivers.

It is not encouraging.

A child born in the bottom fifth of the income distribution has roughly a 7.5% chance of reaching the top fifth. In Denmark, that number is 11.7%. In Canada, 13.4% — approaching twice the American rate. The numbers support what seems like an absurd claim: a child born poor in America has a worse shot at achieving the American Dream than a child born poor in Denmark — or Canada.

Reinforcing the point: a 2020 World Economic Forum report on economic mobility ranked the U.S. 27th.

The generational picture is starker still. Among Americans born in 1940, more than 90% grew up to earn more than their parents. Among those born in 1984, that number had fallen to 50%. Chetty’s team called their paper The Fading American Dream. The title was not rhetorical.

Where you are born matters as much as who you are born to. A child growing up in San Jose has a 12.9% chance of rising from the bottom fifth to the top fifth of the income distribution. In Charlotte, that number is 4.4%. The Dream, it turns out, has a zip code.

Yet economic growth is not the problem. The U.S. economy has roughly tripled in real, inflation-adjusted terms since 1980. America built most of the great companies of the last half century — from Apple to Amazon, Google to Microsoft — and remains the entrepreneurial envy of the world. The problem is not that America stopped producing. The problem is what happened to what it produced.

Over that same period, the bottom half of earners saw their share of national income fall from 20% to 12% — while the top 1% saw their share rise from 12% to 20%. The economy has grown. The growth has not been shared.

Wealth tells the starkest version of the story. According to Federal Reserve data, the top 1% of American households now holds roughly as much of the nation’s total wealth as the entire bottom 90% combined — a level of concentration not seen since 1928, a year before the Great Depression and three years before Adams named the Dream.

F. Scott Fitzgerald was writing about that same era in The Great Gatsby, published in 1925. Gatsby had reinvented himself, accumulated wealth, reached the top. And yet the novel is not a celebration. Fitzgerald dramatized what Adams was about to name: that a society organizing itself around the pursuit of wealth, rather than the expansion of opportunity, produces a different kind of country entirely.

That country is more visible now than at any point since the 1920s.

This is the full shape of the problem. At the bottom, mobility has stalled and the odds of climbing from poverty to prosperity are lower than in many peer democracies. At the top, wealth has concentrated at a rate not seen since the years before the Depression. The ladder still exists. But it is harder to reach, and harder to climb, than at any point in living memory.

“It’s basically a coin flip,” Chetty said, “as to whether you’ll do better than your parents.”

Born in Crisis, Tested Again

Adams coined the term in the depths of the worst economic crisis the country had ever seen.

He also knew it would need defending again.

“Ever since we became an independent nation,” Adams wrote, “each generation has seen an uprising of the ordinary Americans to save that dream from the forces which appeared to be overwhelming and dispelling it. Possibly the greatest of these struggles lies ahead of us.”

He wrote that in 1931.

Each time the Dream has come under pressure — from the Depression, to the long fight for civil rights, to the movements that demanded the promise be extended to everyone it had excluded — Americans have contested it, challenged it, and pushed it closer to what it claimed to be.

The question is not whether that’s possible again. The question is whether we understand what actually broke — and what it would take to rebuild it.

That’s next.

Prefer to listen? I narrate each edition myself. Find the audio at the top of this page or under the Listen tab at solvingfor.io.

Solving For takes on one pressing problem at a time: what’s broken, what’s driving it, and what a path forward might look like. Each series unfolds in weekly installments.

Previous series have examined China's rare earth dominance, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, social media and teen mental health, a world rearming as the global rules-based order weakens, and — most recently — America's national debt crisis. Each series is available for reading or listening (and I narrate them all) at solvingfor.io.

Tristan Harris is the clearest voice we have on what’s at stake in the AI moment. More than a decade ago, he sounded the alarm on social media — long before The Social Dilemma made his warnings mainstream. He thinks we’re repeating the mistake now, at a scale that dwarfs what social media wrought.

We cover a lot of ground: the incentives driving the AI race, the prisoner’s dilemma that keeps it from slowing, the difference between what’s possible and what’s probable, and two primary concerns with more advanced AI models — misuse and loss of control. Harris calls this humanity’s “ultimate test.” Whether AI becomes our greatest tool or our last invention depends on choices being made right now.

If you haven’t read our previous three-part series on AI safety, it’s here. The conversation will make more sense with that context — though it stands on its own.

Much has happened since we recorded this in April. But the framework Harris offers for understanding the stakes hasn’t dated.

A big thank you to Pablo De Ritis, the co-founder and President of Faena Rose, and the entire team.

Looking ahead, our next series examines the American Dream. America’s 250th birthday is approaching, and a Wall Street Journal poll last year found that nearly 70 percent of Americans believe the American Dream no longer holds true — or never did. What’s broken, what’s driving it, and what a path forward looks like. That’s what we’re taking on next.

Previous series have examined China's rare earth dominance, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, social media and teen mental health, a world rearming as the global rules-based order weakens, and — most recently — America's national debt crisis. Each series is available for reading or listening (and I narrate them all) at solvingfor.io.

Show Notes

• 0:00 — Opening by Pablo De Ritis, President of Faena Rose

• 3:24 — Introduction of Tristan Harris by Matt Haggman

• 8:40 — Why AI dwarfs the power of all other technologies combined

• 12:15 — How AI operates: not programmed, grown

• 15:13 — Regulation of AI: “there is more regulation on making a sandwich in New York City than there is for building potentially world-ending artificial intelligence”

• 16:12 — The incentives: thinking about the possible, but not the probable

• 25:55 — The risks: can’t separate the promise from the peril

• 30:40 — Solutions

• 36:30 — Reasons for optimism

• 42:55 — Q&A

Solving For takes on one pressing problem at a time: what's broken, what's driving it, and what a path forward might look like. Learn more here.

In Part Three — the final installment of our series on the national debt — we turn to solutions. How one economist’s warnings are growing harder to ignore. How seven leading think tanks, spanning the ideological spectrum, each produced a credible plan to stabilize the debt. How a Canadian near-collapse in 1995 offers a preview of what may force Washington to act. And why the window for gradual, humane reform keeps closing with every year Congress defers the hard choices.

Missed the earlier installments? Part One explores why the national debt — now larger than the entire U.S. economy — should matter to every American, and why the generation now entering the workforce is the first to inherit a bill they never agreed to run up. Go here. Part Two examines the forces that built it — politicians committing to investments they were unwilling to pay for, shifting the tax burden away from wealth and toward work, and the Baby Boomers whose retirements created a structural deficit no single Congress has been willing to fix. Go here.

Prefer to listen? I narrate each edition myself. Find the audio at the top of this page or under the Listen tab at solvingfor.io.

Jessica Riedl has spent more than two decades doing math that almost no one in Washington wants to see.

She began her career at the Heritage Foundation in 2001, one of the country’s most influential conservative think tanks, analyzing federal spending, deficits, and debt. She later served six years as chief economist to Senator Rob Portman of Ohio, a Republican known for taking fiscal policy seriously in a party that increasingly did not. Today she is a fellow at the Brookings Institution, after eight years at the Manhattan Institute, still following the numbers wherever they lead.

What makes Riedl unusual in Washington is that her conclusions have remained remarkably consistent regardless of political fashion. She has worked in conservative institutions and center-left ones. She has advised Republicans and collaborated with Democrats. Yet her central message has changed very little.

To Riedl, there is no painless solution to America’s outsized national debt.

The gap between what the federal government has promised to spend and what it collects in revenue has grown too large for any single policy to close. It requires a mix across the fault lines of modern U.S. politics: slower growth in Social Security spending, slower growth in Medicare and healthcare spending, less spending across other federal programs, and higher taxes. None of those changes would happen overnight. All would need to be phased in gradually over many years so households, businesses, and retirees could adjust.

The source of the problem is well understood. Americans are living longer, Baby Boomers are retiring, healthcare costs continue to rise, and elected officials have repeatedly chosen borrowing over taxes or spending cuts. Together, those forces have pushed the national debt to 100 percent of GDP — a level the United States has not seen since the years immediately following World War II. The Congressional Budget Office projects the debt will surpass that postwar record this decade.

Subscribe now

Riedl also attacks what she considers budget myths.

Waste, fraud, and abuse are worth addressing, she argues, but the savings are too small to close a fiscal gap measured in trillions. Taxing billionaires more heavily would help, but is not enough by itself. Eliminating foreign aid, shrinking federal bureaucracies, or relying on faster growth may improve the margins, but none fundamentally changes the trajectory.

Riedl has been direct about where that trajectory leads. Unless reforms are enacted, she has written, Washington's escalating borrowing demands will eventually overwhelm the capacity of financial markets to supply lending at plausible interest rates. When that happens, interest rates soar and the federal government cannot pay its bills. The effects would not be abstract — mortgage payments would climb, retirement accounts would lose value, and businesses would pull back on hiring and investment.

In a 2024 interview on PBS’s Firing Line, Riedl noted that when she meets privately with Republicans and Democrats, lawmakers often agree on the broad outlines of what would be required. What they doubt is whether they could survive the politics.

The political will to act, she argues, probably will not arrive “until the sky actually begins falling.”

Economists disagree about the precise mix of spending cuts and tax increases, but there is far less disagreement about the range of options available — and analysts across the ideological spectrum have already produced credible plans to stabilize the debt. Seven of them, in fact.

The obstacle is not analytical clarity. It is creating the conditions under which elected officials are willing to act on what they already know. That is the harder problem — and the one this series ends on.

The Problem Isn’t the Math

How does a democracy fix a problem it has understood for nearly half a century and repeatedly chosen not to address?

The honest answer, supported by the history of nearly every comparable fiscal crisis in democratic governments, is that it usually doesn’t — at least not voluntarily, and not until something forces it.

In 2010, President Obama commissioned a bipartisan deficit-reduction panel led by Democrat Erskine Bowles and Republican Alan Simpson. The commission produced a serious, balanced plan that combined spending cuts and tax increases. It borrowed ideas from both parties and satisfied neither. The commission fell short of the 14 votes it needed to formally advance its recommendations to Congress. The political conditions were not yet desperate enough to make soundness matter.

On the spending side, Social Security and Medicare are unavoidable. Together they represent the largest and fastest-growing share of federal spending. Closing the gap means choosing from a short list of genuinely unpopular options: gradually raising the retirement age, slowing benefit growth for higher earners, reforming healthcare programs to reduce cost growth, or some combination of all three.

Share

The tax side is equally unavoidable. Taxes will need to rise. The debate is over how, and on whom. Possibilities include rolling back the 2017 individual tax cuts — extended permanently by Congress in 2025 — which would raise revenue. Raising the corporate tax rate, which had been cut from 35 percent to 21 percent under the same law, would raise more. Taxing investment income at the same rates as wage income — eliminating a preferential gap that has widened over decades — would raise still more.

Each proposal has supporters and critics. Each carries economic tradeoffs. Each has a well-organized constituency prepared to fight it.

That is the central political challenge. Americans consistently tell pollsters they want the deficit reduced. They also oppose, with equal consistency, most of the specific measures required to reduce it.

The structural problem is not cowardice, exactly. It is rational behavior inside a broken incentive system. Politicians face elections every two or six years. The pain of entitlement reform is immediate, attributable, and politically dangerous. The benefit — a more sustainable fiscal trajectory decades from now — is diffuse and distant.

Politicians are not defying the public. They are following it.

The Solutions Already Exist

In 2024, the Peter G. Peterson Foundation did something unusual. It handed money to seven of Washington’s leading think tanks — organizations that agree on almost nothing — and asked each of them to solve the national debt. Not gesture at it. Solve it.

They called it Solutions Initiative 2024.

The American Enterprise Institute participated. So did the Economic Policy Institute. The Manhattan Institute. The Center for American Progress. The Bipartisan Policy Center. The Progressive Policy Institute. The American Action Forum.

Every single plan worked.

That is the finding worth stapling to the wall of every congressional office. Not one plan — seven. Across every ideological orientation, through combinations of spending cuts and revenue increases that varied in composition, all seven organizations produced a credible path to fiscal stability.

Where the plans agree is striking.

Every organization concluded that healthcare costs are a major driver of long-term debt and must be addressed. Every organization found some changes to Social Security and Medicare unavoidable. Every organization accepted that the gap cannot be closed on one side of the ledger alone.

Where they differ is essentially a values argument dressed in fiscal language.

The right-leaning plans arrive at a smaller government, relying more heavily on spending restraint. The left-leaning plans envision a larger government financed by significantly higher revenues. Both paths stabilize the debt.

The center lands somewhere between those poles, distributing sacrifice across both parties’ sacred cows. A little entitlement reform. A little tax reform. Something for everyone to dislike, which is the definition of a workable compromise.

The obstacle is creating the political conditions under which elected officials are willing to choose.

The Mechanisms for Action

Democracies have developed mechanisms precisely because ordinary politics fails on problems like this one. None is perfect. But they represent the available toolkit.

Bipartisan commissions are the most familiar. Their weakness, as Bowles-Simpson demonstrated, is that they require political leaders to embrace the recommendations after they are made.

A stronger version — modeled on the military base-closing process known as BRAC — would insulate fiscal recommendations from amendment entirely. Under BRAC, a commission's proposals become law automatically unless Congress votes to reject them in their entirety. Lawmakers cannot pick and choose — they can only accept or block the package whole.

Variations of this idea have been proposed repeatedly for Social Security reform, most recently the Bipartisan Social Security Commission Act of 2026, introduced by Republican Tom Cole and Democrat Tom Suozzi. None has passed.

Fiscal rules offer another approach — statutory limits on deficits, debt, or spending designed to constrain elected leaders. The United States has tried versions before. The Gramm-Rudman deficit targets of the 1980s and the budget caps of 2011 both broke down when Congress decided the limits had become more painful than the deficits.

Independent fiscal institutions offer a fourth path. The Congressional Budget Office already plays this role in part by providing nonpartisan scoring of legislation. But the United States has no institution specifically tasked with evaluating government fiscal plans against long-term sustainability targets — and holding leaders publicly accountable when they fall short. The United Kingdom's Office for Budget Responsibility does exactly that. The U.S. does not have an equivalent.

Automatic triggers may be the most technically promising approach. Under such a system, reforms would take effect automatically when a trust fund approached insolvency, unless Congress voted to stop them — flipping the default. Inaction would now produce reform rather than prevent it. Policy experts across the ideological spectrum have proposed versions of this idea.

Each aims to give elected leaders the political cover to make decisions they already know need to be made. The debt keeps rising.

What Usually Forces Action

Riedl has been direct about where the current trajectory leads. “Washington is on a totally unsustainable fiscal path,” she has written, “that virtually ensures some version of a debt crisis.” History offers a sense of what that looks like.

By the early 1990s, Canada had spent years running deficits that were producing familiar warnings: the debt was too large, the trajectory was unsustainable, something had to change.

Cabinet minister Marcel Massé would later recount one such moment when the warnings stopped being abstract. The deputy minister of finance, he wrote, "told me, ashen-faced before a cabinet meeting early in 1994, that 30 minutes before a Government of Canada bond auction that morning there had been no bids.”

In American terms, this would be the U.S. Treasury holding a routine auction for the bonds that finance the federal deficit — and finding no buyers.

Even then, action did not come quickly. Credit-rating agencies were threatening downgrades. International investors were openly questioning Canada's finances. In January 1995, The Wall Street Journal published an editorial calling Canada “an honorary member of the Third World.”

Subscribe now

On February 27, 1995, Finance Minister Paul Martin introduced one of the most aggressive deficit-reduction plans in the country’s modern history. Federal spending was cut sharply. Revenues increased. Three years later, Canada recorded its first budget surplus in nearly three decades.

The argument had been available for years. What changed was that the cost of inaction became immediate and visible. The bond market had delivered a verdict that politicians could no longer postpone.

This is the path many fiscal experts believe is most likely for the United States. Not a grand bargain engineered by statesmen. Not a voter awakening. A market event — a spike in Treasury yields, a failed auction, a high-profile downgrade — that suddenly makes the cost of inaction greater than the cost of action.

Earlier this year, the Committee for a Responsible Federal Budget — a nonpartisan fiscal watchdog whose board includes former senators and cabinet secretaries from both parties — warned that the United States is "woefully underprepared" for the next economic shock. It called on Congress to develop what it labeled a "Break Glass Plan," ready to deploy the moment a crisis strikes.

The CRFB did not say if.

It said when.

Before the Sky Begins Falling

Which brings us back to Jessica Riedl. She has spent years warning that there is no painless path out of America’s debt problem. But there is an important distinction between painful and impossible.

In a 2024 Manhattan Institute blueprint, Riedl proposed gradually raising Social Security's early and full retirement ages to 64 and 69, trimming Social Security and Medicare benefits for higher earners while protecting lower-income seniors, and reforming Medicare through more competitive pricing. On the tax side, her blueprint would restore the top income tax rate to 39.6 percent, remove loopholes to avoid capital gains taxes, and eliminate a tax break the 2017 law created for business owners that salaried workers do not receive. No single element closes the gap. The point is the combination.

Her conclusion mirrors what emerged from the seven Peterson Foundation plans: the details differ, the values differ, the balance between taxes and spending differs — but they all converge on the same point. There is no version of stabilizing the debt that allows Americans to keep every promise, maintain current tax levels, avoid major spending reforms, and still bring the debt under control.

The solutions have been studied, scored, debated, revised, and published for years.

The challenge is political.

The cruelest consequence of delay is that the window for gradual reform keeps closing. Every conventional political promise — “we won’t touch benefits for anyone over 55” — becomes harder to honor with each passing year.

Voluntary reform — the kind Congress could enact today with gradual phase-ins, protections for vulnerable retirees, and decades for households to adjust — would almost certainly be less painful than the version imposed by a crisis. Every year of delay makes the eventual changes larger, more abrupt, and more politically explosive.

History suggests that democracies rarely act until they must. Canada waited until markets forced the issue. The United States may prove different. It may not.

The solutions are written.

The unanswered question is whether America’s elected leaders adopt one before the bond market writes its own.

Prefer to listen? I narrate each edition myself. Find the audio at the top of this page or under the Listen tab at solvingfor.io.

If you enjoyed this look at one of America's most pressing — and most deferred — problems, please give it a like and share with a friend.

Solving For takes on one pressing problem at a time: what’s broken, what’s driving it, and what a path forward might look like.

Previous series have examined rare earth dominance, AI safety, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, social media and teen mental health, and a world rearming as the global rules-based order weakens. Learn more at solvingfor.io.

In Part Two of our series on the national debt, we turn to the forces that built it. How a budget director's confession in 1981 revealed the mechanism that would define the next four decades. How Congress repeatedly lowered taxes on wealth while expanding what the government promised to spend. How 73 million Baby Boomers created a structural deficit no single Congress has been willing to fix. And how rising interest rates may finally be closing the window that allowed Washington to defer the reckoning.

Missed Part One? Go here to understand why the national debt — now larger than the entire U.S. economy — should matter to every American, and why the generation now entering the workforce is the first to inherit a bill they never agreed to run up. Part Three will explore solutions.

Prefer to listen? I narrate each edition myself — audio for this installment is coming soon. In the meantime, Substack's built-in audio is available. All series are at solvingfor.io.

In September 1789, Thomas Jefferson sat down in Paris to write James Madison a letter about money he believed no one had the right to spend.

The city around him was coming apart — the French Revolution only weeks old, the Bastille having fallen in July, the monarchy buckling under debts it could no longer service. Watching the living taxed for debts they had never chosen, Jefferson wrote Madison: “the earth belongs… to the living.” No generation should be able to bind the next with debt it cannot repay within its own lifetime — which Jefferson, working from mortality tables, set at about 19 years.

By 1803, Jefferson sat in Washington as the third president of the United States, and the question he'd put to Madison was about to land on his own desk. Napoleon’s France was willing to sell the Louisiana Territory, and the price was money the young government did not have.

Madison, the man Jefferson warned about the danger of binding future generations, was now his secretary of state. The two men who had argued on paper over whether one generation may mortgage the next found themselves, together, committing the republic to the largest debt of its short life.

Subscribe now

The terms: about $15 million for 828,000 square miles, doubling the size of the country in a single stroke. But the U.S. government did not have the funds in its treasury, nor did America have the capital markets to raise it. The new world turned to the old, issuing bonds through two European banks: Baring & Company in London and Hope & Company in Amsterdam — debt that America would spend two decades repaying, the final installment in 1823. Even the bankers were uneasy. Sir Francis Baring wrote of trembling “at the magnitude of the American account.”

They did it anyway — the purchase encompassed 15 future states, the greatest real estate transaction in history.

The story captures two strands of America's relationship with debt — wary of it in principle, willing to shoulder it when the stakes demanded it.

The Louisiana Purchase was not an exception. America had been born in debt following the Revolutionary War — the first U.S. treasury secretary, Alexander Hamilton, argued that a manageable debt was a national blessing: committing creditors to the new government's success and building the credit the country would need.

Through its history, the country borrowed a great deal — to survive the War of 1812, to preserve the Union through the Civil War, to lift itself out of the Great Depression, and to fight two world wars. World War II pushed federal debt to a record 106 percent of GDP in 1946 — and over the next three decades, the country paid it down to roughly a quarter.

The borrowing had a beginning, a reason, and an end.

Today, that’s no longer the case.

The country is now adding debt at a rate that will eclipse the post-World War II record by the end of the decade, with no end in sight. But there is no war to point to. No depression to climb out of. No continent to acquire. The federal government keeps borrowing without the discipline that once restrained it, or the purpose that justified it.

Deficits Don’t Matter

Today’s debt problem stems from three choices, repeated over the past 45 years until they hardened into habit.

First, politicians committed to things they were unwilling to ask voters to pay for — tax cuts, wars, new benefits. Rather than raise taxes or cut spending, they let borrowing close the gap, knowing voters rarely held them accountable for the deferral. The bill would come later, after they’re gone.

Second, the tax burden increasingly shifted away from wealth and toward work, asking more of the paycheck than the investment portfolio. The shift reduced federal revenues and added to the debt.

Third, Americans lived longer, and the cost of supporting an aging population through programs like Social Security and Medicare continued to climb. What had once been manageable obligations became steadily larger claims on the federal budget.

The roots of the modern era of debt politics can be traced to 1981.

David Stockman, Ronald Reagan’s 35-year-old budget director, spent much of that year arguing that cutting taxes while increasing defense spending would actually shrink the deficit.

Then, in a series of conversations with journalist William Greider that became a damaging profile in The Atlantic that December, he admitted the numbers didn’t add up. Future spending cuts had been replaced by what he called the “magic asterisk”: promises of savings that had never been specified.

The tax cuts were signed. The spending cuts largely were not.

The deficit the asterisk concealed did not disappear. It was simply handed forward.

The details would change. The incentives would not.

There was a brief interruption in the late 1990s, when President Bill Clinton, a Democrat, and a Republican-controlled Congress produced balanced budgets and reduced the debt.

Share

Then the pattern returned.

The 2001 and 2003 tax cuts under President George W. Bush erased projected surpluses. The Afghanistan and Iraq wars were financed through borrowing but, unlike previous wars, without higher taxes or war bonds. America had borrowed to fight wars before. What was new was fighting them while cutting taxes, leaving the bill for future taxpayers. In 2003, Congress added a Medicare prescription drug benefit without identifying how to pay for it.

Former Treasury Secretary Paul O’Neill later recalled that Vice President Dick Cheney dismissed deficit concerns with a simple conclusion from the Reagan years: “deficits don’t matter.”

Borrowing was no longer an exception. It was becoming standard operating procedure.

Then came two genuine emergencies: the financial crisis and the pandemic. Both required massive federal intervention. Both also left the government with a permanently higher fiscal baseline.

Most of the Bush tax cuts were made permanent in 2012 under President Barack Obama, with bipartisan support in Congress. The 2017 Tax Cuts and Jobs Act signed by President Donald Trump cut individual taxes again without offsets and reduced the corporate rate from 35 to 21 percent — the largest such cut in U.S. history. In 2025, Trump and a Republican-controlled Congress extended many of those cuts through the "One Big Beautiful Bill," which the Congressional Budget Office estimated would add $3.4 trillion to the deficit over the next decade.

Debt ceased to be an emergency measure reserved for wars, recessions, or national projects. It became the routine way Washington avoided hard choices.

In 2010, Stockman, a Republican, wrote that the seeds had been planted in 1981, lamenting the “Republican Party's embrace, about three decades ago, of the insidious doctrine that deficits don't matter if they result from tax cuts.”

Taxing Work, Rewarding Wealth

The second shift was quieter than the rise of deficit politics, but just as consequential.

In a 2011 op-ed, titled “Stop Coddling the Super-Rich,” Warren Buffett disclosed that he had paid a federal tax rate of 17.4 percent the previous year. His secretary, Debbie Bosanek, had paid 35.8 percent. The difference was not that Buffett concealed income. It was that most of his earnings came through capital gains and dividends — taxed at lower rates than wages. She was paying more than twice his rate. It was entirely legal.

But the Buffett-Bosanek contrast was not an anomaly. It was the system working as designed.

Over the last 45 years, Congress repeatedly lowered taxes on investment income relative to wages, arguing that lower taxes on capital would encourage entrepreneurship, investment, and economic growth. Today, the top federal tax rate on ordinary income is 37 percent, while long-term capital gains are generally taxed at 20 percent.

The shift changed who carried the tax burden. Wages are taxed automatically through payroll withholding. Investment gains can accumulate for years before being taxed and, when realized, often face lower rates. As asset values rose, more income flowed through forms taxed more lightly than labor.

The full scale of the divide came into view in 2021, when ProPublica published leaked IRS data showing that some of America's wealthiest individuals — including Jeff Bezos, Elon Musk, and Michael Bloomberg — had paid little or no federal income tax in certain years, their wealth surging while taxable income remained small.

The question is simple: why was the government repeatedly lowering taxes on wealth while its commitments — and its debts — kept growing?

The result was not that the wealthy paid no taxes. Many paid enormous sums. The result was that the tax base increasingly leaned on earned income while much of the country’s fastest-growing wealth received preferential treatment.

The government had committed itself to large and growing obligations while simultaneously pulling back from taxing some of the fastest-growing sources of wealth. The gap, once again, was filled with borrowing.

No One Voted for This

The third force is different in character from the first two.

Politicians chose to borrow rather than tax or cut. Politicians chose to tax investment returns more lightly than work. But no politician chose to make Americans older. No one decided that Baby Boomers would retire in waves, that lifespans would lengthen, or that healthcare costs would rise faster than the economy. These things simply happened — and the government’s finances were not built to absorb them.

That was David Walker’s message, and he delivered it to anyone who would listen.

Walker served as U.S. Comptroller General from 1998 to 2008 — the country's chief auditor — and became one of Washington's most persistent voices warning that the long-term fiscal trajectory was worse than most Americans understood.

He took his case on the road in the “Fiscal Wake-Up Tour,” and his warnings became the subject of a 2008 documentary, I.O.U.S.A. Few in Washington were listening.

That was 17 years ago. The debt has since grown fivefold.

The programs at the center of Walker’s alarm — Social Security and Medicare — were designed for a different America. When Medicare was established in 1965, there were roughly four workers for every retiree. The math worked because the young were numerous and the old were not.

That ratio has been narrowing ever since. The Baby Boom generation — 73 million Americans born between 1946 and 1964 — began retiring around 2010 and will continue through the mid-2030s. Lifespans have lengthened. Birthrates have fallen. By 2035, there will be roughly 2.4 workers supporting each Social Security beneficiary, down from four a generation ago.

Healthcare compounds the problem. Medicare costs do not just grow with the number of retirees; they grow with the cost of care itself, which has consistently outpaced inflation and economic growth. An aging population consumes more healthcare, and that healthcare becomes more expensive every year.

The result is a structural deficit that no one voted for and no single Congress created. Social Security and Medicare face a combined funding shortfall of nearly $78 trillion over the next 75 years, according to the programs' trustee reports. The Social Security trust fund is projected to be exhausted by 2033 — seven years from now — at which point benefits would be automatically reduced unless Congress acts. These are not speculative scenarios. They are the official projections of the programs’ own trustees.

This force, unlike the first two, requires no bad faith to explain. The politicians who built these programs were not concealing a time bomb. They were designing a safety net for their moment.

The country changed. The programs largely have not.

This Time May Be Different

Three forces brought the United States to this point: decades of borrowing to avoid difficult choices, a tax system that increasingly favored investment over work, and demographic changes that steadily increased the cost of programs like Social Security and Medicare.

For years, the consequences seemed manageable. The debt grew, but so did the economy. Growth often outran the government’s borrowing costs, allowing Washington to carry ever-larger obligations without an immediate reckoning.

That advantage may be fading, and the reason is interest rates.

Economists have a shorthand for the relationship: r > g, where r is the government's borrowing rate and g is economic growth. When growth is faster than borrowing costs, debt is generally easier to manage. When borrowing costs exceed growth for sustained periods, debt becomes much harder to stabilize.

Interest rates have risen sharply from the ultra-low levels that prevailed for much of the 2010s. The Congressional Budget Office projects that within the next decade the government’s average borrowing costs could exceed the economy’s growth rate — a shift that would make the debt far more difficult to stabilize and increase the risk of a debt spiral.

That prospect has begun to change minds.

In a May 2026 essay for The Atlantic, economist Jared Bernstein noted that he had spent much of his career dismissing claims that a high debt-to-GDP ratio signaled imminent crisis. But if interest rates begin to outpace growth, he argued, the calculus changes. “If you’re not worried about this country’s fiscal outlook,” Bernstein wrote, “you’re not paying enough attention.”

Jefferson’s warning more than 230 years ago was about more than debt. It was about accountability — the idea that the people who make a decision should be the ones who live with it. For most of American history, that principle held, imperfectly but recognizably: wars were fought and paid for, crises met and resolved. What the past 45 years produced is something different — debt without emergency, spending without payment, commitments made by one generation and quietly handed to the next.

At some point, there is no one left to pass it to.

Prefer to listen? I narrate each edition myself. Find the audio at the top of this page or under the Listen tab at solvingfor.io.

If this resonated, please give it a like and share with someone you think would enjoy it.

Solving For takes on one pressing problem at a time: what’s broken, what’s driving it, and what a path forward might look like.

Previous series have examined rare earth dominance, AI safety, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, social media and teen mental health, and a world rearming as the global rules-based order weakens. Learn more at solvingfor.io.

U.S. government debt has been a fixture of American life for decades. And it hasn’t seemed to matter. Markets keep humming. The economy keeps growing. Life goes on.

Yet, the national debt recently surpassed the size of the entire U.S. economy. The federal government is on its way to setting a new all-time record for indebtedness, and there is little end in sight. Now, some are sounding alarms they haven’t sounded before about the risks it presents today, and the burden it places on the next generation.

Today, we explore the problem and why it should matter. Next week: what’s driving this. The week after: what solutions might actually look like. Every Solving For series is available to read or listen to — I narrate each one myself —at solvingfor.io.

Robert Giaimo and Henry Bellmon had almost nothing in common.

Giaimo was the son of an Italian immigrant banker, raised in New Haven, Connecticut, educated at Fordham and the University of Connecticut law school, a Democrat who spent two decades in the House representing the factory workers and union households in southern Connecticut. Bellmon was born on a farm near Tonkawa, Oklahoma, the son of a Cherokee Strip pioneer, a man who worked his way through Oklahoma A&M College, saw battle as a first lieutenant at Saipan and Iwo Jima, and served two terms in the U.S. Senate, repeatedly returning to his wheat and cattle farm in Noble County. He was a Republican, but he considered himself, above all else, a farmer.

In 1980, both men left the U.S. Congress. They had spent years on opposite sides of the House and Senate Budget Committees — Giaimo as chairman in the House, Bellmon as the ranking Republican in the Senate — watching from the inside as the federal government found it consistently easier to spend than to choose. They had seen the oil shocks, the stagflation years, the beginning of what would become a structural mismatch between what the government promised its citizens and what it was willing to collect from them.

When they left, rather than simply returning to private life, they did something unusual: they decided the problem required an institution. On June 10, 1981, they joined together to incorporate the Committee for a Responsible Federal Budget — a nonpartisan organization devoted to the proposition that the country's fiscal path was unsustainable and that someone, outside the political pressures of government, needed to keep saying so.

Nearly 45 years later, the organization's president, Maya MacGuineas, explained how far the problem had traveled.

“It’s happened,” she declared in April 2026. “The national debt is now larger than the U.S. economy, about twice the historic average. We’ve heard plenty of alarm bells in the past few years about our fiscal path, but this one rings especially loudly.”

The Congressional Budget Office projects that by the end of the decade the U.S. will set a new all-time record for debt held by the public — surpassing the 106 percent of gross domestic product reached in the immediate aftermath of World War II.

Unlike that debt, which the country reduced systematically over the following three decades, this one has no equivalent plan behind it — just a projection that keeps climbing. And there is a big difference, MacGuineas noted, between that moment and this one. The post-World War II debt was the price of defeating fascism. This one, she said, is the result of “a total bipartisan abdication of making hard choices.”

Why It Matters

This three-part Solving For series explores America’s national debt: why it matters, the forces that got us here, and what a way forward could look like.

It's a problem that has long escaped serious attention or any meaningful reform. But the consequences are no longer theoretical. Continued inaction carries risks to the economy, to American competitiveness, and to a generation that will inherit the bill without having agreed to run it up.

The interest payments on the debt are now the federal government's second largest spending category — more than national defense or Medicare, trailing only Social Security. Within a generation it's projected to become the federal government's single biggest expense, crowding out every public investment that historically drives prosperity: education, infrastructure, research, healthcare.

Today, nearly one-fourth of those interest payments flow to foreign countries, including China, “to build their economies rather than our own,” according to RAND, one of the country’s most respected national security and policy research institutions.

Meanwhile, the debt makes borrowing more expensive for every American who needs a mortgage, a car loan, or a business line of credit. The reason: government borrowing at this scale competes with every other borrower in the same capital markets — and it always wins, driving up the cost of credit across the entire economy.

Subscribe now

It shrinks the fiscal space available to respond to the next crisis. 2008 brought a Great Recession; 2020 brought COVID-19 and a global pandemic. There will be another.

It puts Social Security and Medicare on an insolvency countdown that, without action, ends in automatic benefit cuts for those who need them most.

At a time when the gap between the wealthiest Americans and everyone else has reached the widest point in decades, it dramatically limits the government's capacity to find solutions.

And it carries the risk of a sudden loss of market confidence that turns an outsized debt burden into a full-blown financial crisis.

It’s so concerning that — despite structural advantages like the U.S. dollar's reserve currency status, the depth of American capital markets, and the scale of the U.S. economy — Hank Paulson, former Treasury Secretary under George W. Bush, warned in April: “We need an emergency break-the-glass plan…so it’s ready to go when we hit the wall.”

When. Not if. And when it comes, he said, “It will be vicious.”

But the deepest problem isn't the number. A government that spends more on interest payments for past decisions than investments in future growth is one that has stopped making choices — and started living with the consequences of decisions it can no longer take back.

The World’s Richest Debtor

To understand the problem of America’s national debt, start with the economy itself.

The United States has the largest economy in the world — by far. As of the close of the first quarter of 2026, American GDP stood at $31.22 trillion — more than one quarter of all global economic output. For context: China is second at $19.4 trillion, Germany a distant third at just over $5 trillion, and Japan fourth at $4.28 trillion. The U.S. has been the most productive economic engine in human history for more than a century.

And yet. As of March 31 of this year, federal debt held by the public stood at $31.27 trillion — larger than the largest economy in the world. That figure refers to what the government owes outside investors, as distinct from the gross debt figure of $39 trillion, which includes money the government owes to itself.

The debt grows because the federal government continues to run large annual budget deficits — the yawning gap between what it collects in revenue and what it spends each year. To cover the difference, the government borrows by issuing Treasury bonds, compounding the debt.

Consider 2025. The federal government collected $5.2 trillion in revenue and spent $7.0 trillion. The $1.8 trillion gap was borrowed and added to the debt.

It's an especially alarming case of financial mismanagement, but not an anomaly. With a few exceptions, it's become the operating condition of the United States government, sustained across administrations and Congresses of both parties for decades.

The deficit is a two-sided problem. The spending side gets most of the attention. The revenue side is equally revealing — and equally the result of deliberate choices made over decades. Part Two will examine those choices, including a consequential and long-running debate about how investment income is taxed compared to wages — particularly for the wealthiest Americans.

The 2025 budget illustrates the basic picture: where the money comes from, and where it goes.

Of $5.2 trillion in tax revenues, just over half came from individual income taxes on both wages and investment gains — two forms of income the tax code treats very differently; roughly one-third from payroll taxes that fund Social Security and Medicare; about seven percent from corporate income taxes; and the rest from a mix of excise taxes, tariffs, and fees.

The $7 trillion in federal spending falls into three categories, and the proportions reveal something important about the country's actual priorities — not as stated, but as enacted.

The first and largest is mandatory spending: programs governed by permanent law that run on autopilot unless Congress specifically acts to change them. Mandatory outlays totaled $4.2 trillion in 2025 — more than half of the entire federal budget. Social Security, the retirement and disability program, costs roughly $1.6 trillion a year. Medicare — the federal health insurance program for Americans 65 and older — accounts for nearly $1 trillion, and Medicaid — health insurance for low-income Americans of all ages — another $670 billion. These are not discretionary line items subject to annual negotiation. They are legal promises made to tens of millions of Americans who have spent their working lives paying into the system.

The remaining roughly $900 billion in mandatory spending funds other legally mandated commitments: veterans' benefits, federal employee retirement, student loan programs, and agricultural subsidies.

Share

The second is discretionary spending — programs Congress funds through annual appropriations, subject to yearly negotiation. In 2025, discretionary spending was just 27 percent of the total budget. Defense accounts for the largest share of that — roughly $900 billion. The Trump administration is seeking to increase that figure to $1.5 trillion in the coming fiscal year. Everything else that most people associate with the federal government — education, transportation, scientific research, environmental protection, housing, foreign aid — competes for what remains.

When politicians talk about cutting government spending without touching entitlements or defense, they are negotiating over a relatively small fraction of the whole.

The third is interest on the national debt. In 2025, that totaled nearly $1 trillion — $2.8 billion every day. Interest has become the fastest-growing line item in the entire federal budget. No roads, no research, no benefits, no security. It is the pure cost of decisions already made, extracted from the present.

Stated plainly, the federal budget reflects a specific set of national priorities that have been locked in across decades of legislation: supporting the retirement income of older Americans, funding healthcare for the elderly and the poorest Americans, maintaining a global military presence, and increasingly, servicing the debt accumulated while pursuing all three. Everything else — the investments that tend to determine a country’s future — is funded from what remains after those obligations are met.

In 2025, that remainder was not enough to cover the obligations themselves.

The Generation That Didn’t Vote for This

Researchers at the Harvard Kennedy School conduct what many consider the gold standard of American youth polling — a nationally representative survey of young adults ages 18 to 29, now in its 51st edition. The findings released in December 2025 told an unambiguous story. More than four in ten young Americans said they were struggling or getting by with only limited financial security. A clear majority said the country was on the wrong track. Money was at the root: the sense that the economic system their parents described to them is not the one they actually inherited.

Behind that instability is a specific and measurable inheritance. The generation now entering its working years is the first to do so inheriting a national debt larger than the entire economy — accumulated across decades of decisions made entirely by others. They did not vote for the tax cuts. They did not vote for the unfunded wars. They did not vote for the entitlement expansions or the emergency stimulus programs or the structural mismatch between what the government promised and what it was willing to collect. They are, in the precise sense, inheriting a bill they never agreed to run up.

The people bearing the consequences are not the people who made the decisions. That is not an accident. It is the natural outcome of a democratic system in which the costs of present decisions are deferred long enough that those who will pay them cannot yet vote.

Robert Giaimo and Henry Bellmon saw this coming in 1981. They built an institution to say so, and that institution has been saying so for nearly 45 years. The alarm has not changed. The distance between where it was first sounded and where we now stand has. The debt has crossed 100 percent of GDP. Interest payments have surpassed defense spending. A former Treasury Secretary is calling for an emergency break-the-glass plan. And a generation is entering a world shaped entirely by decisions made before they arrived, by leaders who will not be here to answer for them.

The bill, as bills sometimes do, has found its way to the last people left at the table.

Prefer to listen? I narrate each edition myself. Find the audio at the top of this page or under the Listen tab at solvingfor.io.

If this resonated, please give it a like and share with someone you think would enjoy it.

Solving For takes on one pressing problem at a time: what’s broken, what’s driving it, and what a path forward might look like.

Previous series have examined rare earth dominance, AI safety, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, social media and teen mental health, and a world rearming as the global rules-based order weakens. Learn more at solvingfor.io.

We return to The Open Thread, our periodic check-in on past stories and a look at what’s ahead. Today we revisit four of them: A.I. safety landed on the diplomatic agenda at the U.S. and China summit, the fight for competitive congressional elections has gotten much harder, local news got a rare break, and college sports' new system governing how student athletes are paid is beginning — albeit shakily — to show its shape. All series (for reading or listening — I narrate each one) are at solvingfor.io.

AI Safety: Reconsidered

When we covered A.I. safety last fall, the central tension was this: the U.S. and China were locked in a race that neither side felt it could slow down, even as the creators of A.I., top CEOs and leading researchers warned of the risks of moving too fast. The prisoner’s dilemma of A.I., where mutual caution is the rational choice but unilateral restraint feels like surrender, seemed nearly impossible to escape.

The solution we highlighted: the U.S. and China sitting down to negotiate ways to manage the risks of A.I., just as the U.S. and Soviet Union did nearly 60 years ago (at the height of the Cold War) to control the growth and spread of nuclear weapons.

Last week, we may have seen the start of it.

Ahead of the summit between Trump and Xi in Beijing, both The Wall Street Journal and The New York Times reported that A.I. safety would be on the agenda. It came as the Trump administration had done an about-face on the need to explore guardrails, as frontier A.I. models by the likes of Anthropic, Google DeepMind, and China’s DeepSeek grow more powerful on both sides of the Pacific.

The previous posture was clear. Vice President J.D. Vance gave a speech last year in Paris declaring that the “A.I. future is not going to be won by hand-wringing about safety.”

Subscribe now

Why the change? The catalyst is Anthropic’s powerful new model, Claude Mythos, whose advanced hacking capabilities spooked government officials and set off a scramble about what the government’s role should be in managing the release of the most powerful AI systems. The White House has since moved toward considering pre-deployment safety testing — a sharp reversal from its earlier position of prioritizing rapid innovation without any guardrails.

But the fear remains: any slowdown by one side gives the other an advantage. Hence the push for a global framework established by today’s great powers. Not only rules governing the release of ever-more-powerful A.I. models — systems that could pierce financial networks or assist in the development of bioweapons — but procedures to manage a crisis. Proposals include establishing an emergency hotline between the U.S. and China for A.I. matters, modeled on the one the U.S. and Soviet Union created during the Cold War.

“The two A.I. superpowers are going to start talking,” U.S. Treasury Secretary Scott Bessent said during the meetings in Beijing. “We’re going to set up a protocol in terms of, how do we go forward with best practices for A.I. to make sure nonstate actors don’t get ahold of these models.”

Following the summit, President Trump was circumspect about specifics, but confirmed the change in approach. “AI is fantastic,” Trump said aboard Air Force One, according to The Hill. But “it’s also got some drawbacks and we’re talking about…we’re going to work together.”

A protocol. A hotline. The echoes of the nuclear age are getting harder to ignore.

If you missed it, you can explore the three-part series, The Control Problem: Solving For AI Safety, here.

Congress: Not a Competition of Ideas — Just Power

Start with the baseline: in 2024, just 37 of 435 House races were decided by five points or less — nine percent. Most Americans now live in districts where the outcome is effectively predetermined long before Election Day. The consequence: rising polarization, a dearth of genuine debate, and a Congress no longer structured to reward solving problems — only holding power.

The House was designed by the Framers to be the chamber closest to the people — directly elected, up for election every two years, meant to mirror public sentiment. It has become the opposite.

When we covered the alarming lack of congressional competition, the core problem was structural: the way districts are drawn drains Congress of the competitive pressure that produces accountability to the entire electorate. Closed primaries compound it, pushing candidates toward the extremes before a general election audience ever weighs in. The solutions we examined — including independent redistricting commissions, open primaries, and ranked-choice voting — were slow and contested, but at least still in play.

In recent weeks, the landscape has gotten dramatically worse — and it is worth being clear about how we got here.

Facing increasingly dire polling numbers last summer, President Trump urged Republican-led states to gerrymander their congressional districts to boost the party’s chances in the upcoming midterms. It was a politician choosing his voters rather than voters choosing their politicians — and an extraordinary departure from tradition: congressional redistricting typically follows the decennial census, not a president’s approval ratings. Texas, North Carolina and Missouri quickly complied. Florida recently did too. It was nakedly political — and it set the stage for what followed.

The legal architecture enabling it came in two steps. In Rucho v. Common Cause, the Supreme Court ruled in 2019 that partisan gerrymandering is a political question — federal courts would stay out entirely. Then on April 29th, the Court went further in Louisiana v. Callais, effectively gutting the one remaining constraint: the Voting Rights Act’s protection of minority voters. Justice Kagan, in dissent, warned the majority had rendered Section 2 of the Voting Rights Act “all but a dead letter.” Together the two decisions form a closed loop — partisan gerrymandering is unreviewable, and race can no longer meaningfully constrain it.

The human cost is concrete. About one-third of Louisiana's population is Black, yet under the new map being advanced, the state would go from two majority-Black congressional districts back to one. The partisan math is equally stark: Louisiana has nearly identical numbers of registered Democrats and Republicans — and will now likely send five Republicans and one Democrat to Congress.

Share

The ruling immediately opened a new front. Tennessee drew a new map intended to eliminate the state's last Democratic House seat. South Carolina's governor called the legislature into special session to pursue a redraw. In Alabama, the attorney general was candid about his goal: a congressional map that "favors Republicans seven-to-zero." In a state where 27% of the population is Black — and where Black voters overwhelmingly vote Democratic — eliminating Black representation and maximizing Republican advantage are a single goal.

Democrats have tried to fight back. California voters passed a new map adding five Democratic-leaning districts. Virginia voters approved a redistricting referendum, only to have the Virginia Supreme Court strike it down 4-3, leaving the state's current maps in place through the midterms.

The point is this: the effort to make Congress more competitive, more representative, and more accountable has not stalled — it has been deliberately dismantled.

The remedies we explored still exist on paper: a federal ban on partisan gerrymandering, independent redistricting commissions, open primaries, multi-member districts, and ranked-choice voting all remain possibilities. But the wheels for reform have completely come off. What’s left is the rawest version of the problem we identified — not a competition of ideas, but a competition for power, with the maps drawn accordingly.

If you missed it, you can explore the three-part series, The Democracy Deficit: Solving For Competition in the People’s House, here.

Local News: A Model Worth Watching

When we covered the collapse of local news, the story was grim: newsrooms gutted, communities left without watchdogs, and no sustainable business model emerging to fill the void at scale.

Here’s a piece of good news.

The Pittsburgh Post-Gazette, a 240-year-old newspaper set to close, was purchased by the Venetoulis Institute for Local Journalism — the nonprofit parent of The Baltimore Banner — which took ownership May 4th and pledged $30 million over five years to keep it running. The Venetoulis Institute was established by Baltimore billionaire Stewart Bainum after he failed to acquire The Baltimore Sun. Instead, he launched the Baltimore Banner from scratch in 2022 — a digital-only nonprofit that won the Pulitzer Prize for local reporting. This isn’t a philanthropic rescue with no model behind it. It’s an organization that has already proven nonprofit journalism can produce serious work.

The Post-Gazette would become only the second major metro daily to fully convert to nonprofit, following The Salt Lake Tribune. One rescue does not reverse a national crisis. But Pittsburgh — which had been on the verge of becoming the largest American city without a daily newspaper — now has a model worth watching.

If you missed it, you can explore the three-part series, The Community Gap: Solving For Local News, here.

Paying College Athletes: Still Taking Shape

When we covered the end of amateurism in college sports, the headline was simple: the old system was dead. What remained was the harder question — how do you fairly compensate athletes who generate billions while preserving the traditions and educational mission that give college sports its meaning; and fund the non-revenue Olympic sports like track and field that football and basketball have historically bankrolled?

The system is a work in progress. But it has structure now, and that structure recently faced its first real test.

At the center of it is the House settlement, reached last year, which established a $20.5 million per-school pool for direct revenue sharing between schools and athletes — policed by a newly created body called the College Sports Commission. But the richest programs have been spending well beyond that cap. An ESPN report said the most competitive college football programs spend between $30 to $40 million to assemble their rosters. The mechanism: name, image and likeness — NIL — deals structured through third-party marketing firms, which sit outside the cap entirely. The CSC’s position is that those deals have to represent genuine commercial value — a company paying for specific, identifiable use of a player’s likeness — not a backdoor salary payment in disguise.

Last week, an arbitrator upheld the CSC’s decision to deny NIL deals involving 18 Nebraska football players and Playfly Sports, Nebraska’s multimedia rights partner, which had pledged more than $8 million to athletes through arrangements ruled to be “warehousing” — purchasing NIL rights without specifying how they’d actually be used. The ruling signals the new system has teeth. But it is less an ending than a first skirmish: a federal court hearing is scheduled for May 27 to review the CSC’s authority over exactly these kinds of deals, and Nebraska’s attorney general has signaled he may intervene.

A system taking shape. The tests are just beginning.

If you missed it, you can explore the three-part series, The Amateur Myth: Solving For College Athlete Pay, here.

Next up

For our next series, we turn to the U.S. federal debt. The numbers alone are stark: the debt now exceeds the entire U.S. gross domestic product — a milestone that is casting a harsh new light on how the government taxes, spends, and borrows. We’ll unpack the problem, trace how we got here, and examine what might actually be done about it.

Note: Prefer to listen? I narrate each story myself. Use the Article Voiceover at the top of the page, or find all narrated editions in the Listen tab at solvingfor.io.

Solving For is a deep-dive series that takes on one pressing problem at a time: what’s broken, what’s driving it, and what a path forward might look like.

Previous series have examined rare earth dominance, AI safety, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, and a world rearming as the global rules-based order weakens. Learn more at solvingfor.io.

To see the full roster of deep dives, take a look below —

The 21st Century’s Oil: Solving For China’s Rare Earth Dominance

• Part 1, Rare Earths: The Invisible Backbone

• Part 2, Rare Earths: The Middle Kingdom’s Monopoly

• Part 3, Rare Earths: The Race to Reinvent

The Control Problem: Solving For AI Safety

• Part 1, AI: The Race and the Reckoning

• Part 2, AI: The Prisoner’s Dilemma

• Part 3, AI: The New Nuclear Moment

The Democracy Deficit: Solving For Competition in the People’s House

• Part 1, Congress: The Vanishing Competition

• Part 2, Congress: How We Got Stuck

• Part 3, Congress: Making Democracy Competitive Again

The Amateur Myth: Solving For College Athlete Pay

• Part 1, College Sports: How It Was Broken By a $60 Video Game

• Part 2, College Sports: How the NCAA was Born of Death and Money — Death was the Easy Part

• Part 3, College Sports: The Fork in the Road

The 80-Year Peace: Solving For a Rearming World

• Part 1, War & Peace: The World Rearms

• Part 2, War & Peace: From 1989’s Hope to Rearmament

• Part 3, War & Peace: Rebuilding the Global Order

The Community Gap: Solving For Local News

• Part 1, Local News: The Civic Unraveling

• Part 2, Local News: The Internet Was the First Disruption. AI Is the Next.

• Part 3, Local News: The Bet That Wasn’t Made, But Could Still Be

Engineered Addiction: Solving for Teen Mental Health

• Part 1, Teen Crisis: How Facebook Targeted Tweens

• Part 2, Teen Crisis: The Twenty-Six Words that Shielded Social Media

• Part 3, Teen Crisis: Three Ways Forward

In Part Three of our series on social media’s impact on teen mental health, we turn to solutions. How accountability is being pursued on two fronts at once — through more than a thousand cases working their way through the courts in the opening created by the KGM v. Meta verdict, and through efforts like forty state attorneys general pushing Congress to write a duty of care into law. How Australia raised the minimum age to have a social media account and a growing number of countries are following. How parents, schools, and kids themselves stopped waiting for Washington and went phone-free. And how the question of whether the next generation grows up under different terms is now being answered.

Missed Part One? Go here to unpack the epidemic of depression, anxiety, and, tragically, suicide among a generation of teenagers — and social media’s role in it. Missed Part Two? Go here for how a 1996 law shielded the industry, how the platforms engineered the harm behind that shield, and how lawyers built a legal theory to get around that shield. All series (for reading or listening) are at solvingfor.io.

By the time Jonathan Haidt and Catherine Price walked onto The Daily Show set on the night of February 25, 2026, the cultural movement they had helped build was at its most visible. They were in midtown Manhattan to explain to Desi Lydic and her studio audience what had happened to American childhood — and what was, finally, beginning to happen in response.

Haidt had written The Anxious Generation, the 2024 book that crystallized a global conversation about the effects of smartphones and social media on young people's development. Price was the author of How to Break Up With Your Phone. Together they had written a new book for kids called The Amazing Generation. Haidt was on the show to talk about both — but mostly to talk about what had happened in the nearly two years since The Anxious Generation had landed.

A great deal had happened.

“In the first year, we got 40 US states to put limits on phones,” Haidt told Lydic. “Twenty did it right, phone-free for the whole day. Australia is the first country in the world to raise the age to 16 for social media.” The audience cheered. “And just in the last four weeks, a dozen countries have said they're going to follow. So we are at a global turning point.”

The cultural movement Haidt had helped launch — parents organizing in PTA meetings and statehouses, schools rewriting their phone policies, governments passing laws that would have been unthinkable two years earlier — was, by any measure, ascendant.

Subscribe now

Then Lydic asked the question the entire movement had been built around.

"How much of the onus falls on the parent and the teachers," she said, "and how much falls on these tech companies and our leaders to do the right thing?"

Haidt's answer is worth quoting, because everything that follows in this installment is, in one way or another, a response to it.

“In any rational world where you had a consumer product that was used by 95% of all children, that had killed thousands of them, that had gotten to the point where 25% of the girls say that it has damaged their mental health…it would have been sued out of existence or at least made an adult-only product.”

“Of course, we don’t live in that world,” Haidt continued. “We live in a world where wealthy industries can just pay not even that much money to buy huge amounts of influence in Congress and block anything.”

But, he said, the states were acting. The rest of the world was acting too. “I think we're going to win on this eventually, and the companies are going to have to bear some responsibility. Of course, in Los Angeles right now, they're on trial for the first time. They're facing a jury. Thousands of kids are dead and they've never had an answer for it.”

“So I think justice will come,” he concluded.

Twenty-eight days later, it did.

On March 25, 2026, in downtown Los Angeles, the jury in KGM v. Meta et al. found that Meta — owner of Facebook and Instagram — and YouTube had breached a duty of care to the minors who used their products. The verdict was the first of its kind: an American jury had held a major social media company liable, on a product-design theory, for harm to a child.

The legal architecture that had shielded the platforms for nearly thirty years had cracked. The rational world Haidt had described had begun to arrive.

For Part Three, we turn to solutions. Since adolescent depression, loneliness, and suicide began their rise after the smartphone and the always-on social platform reshaped American childhood, the work of finding a fix has been underway. It has broadly coalesced around three approaches.

Fix the product. Hold the platforms accountable for the design of what they sell — through litigation like KGM, through legislation like the Kids Online Safety Act, and through regulation requiring age-appropriate design codes.

Raise the age. Restrict access until children are older. Australia's age-sixteen law took effect in December 2025. A growing number of countries are following.

Change what’s normal. Cultural change among parents, schools, and kids themselves. Phone-free school days. Parents organizing in PTA meetings to break the collective action trap that keeps individual families from saying no. Norms about when children get smartphones and when they get social media. Helping kids make their own good decisions.

“This all makes so much more sense if you stop focusing on the phones and you focus instead on childhood,” Haidt said on The Daily Show. “What is a healthy human childhood?”

Fix The Product

The most direct response to the harms documented in this series is the one the KGM jury delivered: hold the platforms accountable for the design choices that caused the harm. This is the project of taking that verdict and making it the rule rather than the exception.

There are two ways to do this: civil litigation and legislation.

Civil litigation rests on a single legal doctrine: duty of care. It's the long-standing principle that anyone who designs and sells a consumer product owes its users a basic obligation not to design it in ways that foreseeably injure them.

For thirty years, platforms operated outside this framework, because Section 230 was read to immunize them from claims arising out of the content their users posted. The KGM verdict did not overturn Section 230. It established something narrower but consequential: that design-defect claims targeting the platform’s own conduct — its recommender systems, its notification engineering, its retention features — survive Section 230 because they are not about user content. They are about the product.

The path applies duty of care case by case. The cases are piling up. Behind the KGM jury verdict sits a docket that increasingly resembles the early years of tobacco litigation: more than a thousand personal injury cases against Meta, YouTube, Snap, and TikTok consolidated in California state court, and more than two thousand coordinated in federal court.

The social media companies have begun settling some cases and fighting others, the same posture Big Tobacco took in the early 1990s before the cumulative weight of verdicts and disclosures produced the 1998 Master Settlement Agreement. Whether the same arc holds for social media remains an open question. What is clear is that the legal wall that protected the industry for thirty years has cracked, and the litigation is moving through the opening.

Share

Legislation takes a different route: write the duty of care into statute. This approach is illustrated by the proposed Kids Online Safety Act, which would set the standard in federal law and let the Federal Trade Commission and state attorneys general enforce it across the industry. KOSA passed the Senate 91-3 in July 2024. The House version, advanced by Republican leadership, stripped the duty of care entirely. Forty state attorneys general wrote in February 2026 backing the Senate version, but the legislation is stuck.

In the absence of federal action, states have moved. Age-appropriate design codes have been enacted in California, Maryland, Nebraska, Vermont, and South Carolina, imposing requirements like privacy by default, no targeted advertising to minors, and a ban on dark patterns such as infinite scroll, autoplay, and streak counters.

Several of these laws are now under First Amendment challenge. In NetChoice v. Bonta, the Ninth Circuit affirmed a preliminary injunction against parts of California's law. Civil liberties groups argue that a duty of care covering the content kids see — not just how the platform is built — chills lawful speech, particularly speech about mental health, sexual orientation, and reproductive care that vulnerable teens often most need to find.

But even if every mechanism outlined here succeeded — if KOSA passed with the duty intact, if the design codes survived constitutional challenge, if KGM became the leading edge of a thousand similar verdicts — the question of whether children should be encountering these products at all would remain.

Fixing the product makes the product safer. It does not answer when, or whether, a child should be using it.

Raise The Age

The second approach starts from a different premise: until the product is safer, keep children away from it.

Australia’s age-sixteen law took effect on December 10, 2025. It applies to Facebook, Instagram, TikTok, Snapchat, YouTube, and X — the platforms most American teenagers use, the platforms named in the KGM litigation, the platforms a generation of parents has watched their children disappear into. Within the first six weeks Australia’s eSafety Commissioner reported that 4.7 million accounts belonging to under-sixteens had been removed, with non-compliant platforms facing fines of up to AUD$49.5 million. Greece has announced that a ban at 15 will start in 2027. France is debating a similar law. Indonesia, Brazil, Austria, Denmark, Malaysia, Ecuador, and others have moved or are moving in the same direction.

The cultural and political logic is straightforward. Individual families could not keep their children off the platforms because the platforms were where their children’s friends were — the collective action trap that Haidt and others have spent two years documenting. A government restriction breaks the trap by making the platform legally inaccessible to everyone at once.

Subscribe now

On this question, at a time of deep political polarization in the U.S., the usual partisan map is not holding. Congress has not moved, but states have, and the alignments are unfamiliar. Republican and Democratic attorneys general have filed the same suits. Red-state governors and blue-state senators have backed the same restrictions. Florida’s Republican Governor Ron DeSantis, who signed a law banning Floridians under 14 from holding social media accounts, said, “Being buried in those devices all day is not the best way to grow up.” Connecticut’s Democratic Senator Richard Blumenthal is the lead sponsor of KOSA.

The opposition is bipartisan too: the ACLU on speech and privacy grounds, libertarian conservatives on parental rights. When neither coalition maps onto the usual axis, it may signal something is about to give.

It is too early to tell whether age restriction reduces the harms it targets. Australia’s law has been in effect for less than six months. Public health interventions of this kind take a decade or more to show up clearly in outcome data. What the early reports do show is that the law is not yet reaching the kids it's meant to protect: by March 2026, the eSafety Commissioner found no measurable decline in reported cyberbullying or image-based abuse, and roughly 70 percent of parents whose children had previously held accounts believed their children still did.

The deeper questions are about the framework itself. Candice Odgers at UC Irvine and Amy Orben at Cambridge have long argued that the causal evidence linking social media to youth mental health decline is weaker than the Haidt framework presents, raising doubts about whether interventions built on separating teens from social media will deliver the outcomes proponents promise.

But the case for raising the age does not depend on age restriction being sufficient on its own. It buys time during the years when adolescent brains are still forming. It breaks the collective action trap from above, in a way no individual family can break it from below.

Raising the age does not fix the product, and it does nothing for the older teenagers already on the platforms. But it changes the terms on which the next generation encounters them.

Change What’s Normal

The third approach starts from a different question. Even if every mechanism in the first two approaches succeeded — accountability forcing the products to be safer, every age line raised — children will still encounter them. The question is what kind of children encounter them, and what tools they bring to the encounter.

This is the work of cultural change. It is the work parents, schools, and kids themselves can do without waiting for Congress, the courts, or any government at all. It moves at the speed of families talking to other families, of school boards or a single school principal rewriting their phone policies, of friend groups deciding together what they will and will not do. It does not require legal force.

The most widely adopted intervention is the phone-free school day. By early 2026, 35 states had enacted or recommended classroom phone restrictions, and many districts and individual schools had moved on their own ahead of state action. The early data is encouraging: a study of Florida's statewide ban found rising test scores and falling discipline problems by the second year. At one Louisville high school, library checkouts jumped 61 percent after phones came out of the classroom, and the library became, in the principal's words, a place where students "look at each other in the eyes and talk." At Portland's Grant High School, the principal noticed something simpler after his school went phone-free: lunchtime, he said, was loud again.

Alongside this sit broader efforts: Wait Until 8th, which asks parents to delay giving their children smartphones until at least eighth grade, and similar pledges that let parents act in coordinated cohorts rather than as isolated families. There is also a growing body of writing aimed at kids and teens themselves, including Catherine Price and Jonathan Haidt's The Amazing Generation, which is framed as an invitation rather than a prohibition.

Students are taking matters into their own hands. Student-led efforts like the Reconnect Movement, which creates phone-free social spaces without asking participants to give up phones entirely, have been spreading on high school and college campuses.

Whether phone-free schools and parent pledges reduce adolescent depression and anxiety at the population level is not yet known, and cannot yet be known. The phone-free school day movement is barely two years old at scale.

But the cultural-change approach does something the other two cannot. It does not protect children from a world that includes these platforms; it prepares children for that world. Phone-free school days are not just restrictions — they are practice in what attention without a phone feels like. Parent pledges are not just delays — they are demonstrations of what coordinated adult judgment looks like. The youth movement Haidt and Price describe is more than refusal — it is kids exercising the judgment they will need for the rest of their lives.

The goal is not a smaller world for children. It is the capacity to live in this one.

The Work Has Begun

What is a healthy human childhood?

The harms documented in this series did not arrive overnight, and the answer to them will not arrive overnight either. The tobacco settlement took decades to assemble. The leaded-gasoline ban took a generation. Public smoking bans came long after the science was settled. Industrial harm to children, once it has been built into a profitable consumer product, does not unwind on a political news cycle.

The lesson of the years since Section 230 became law is that this is not a decision the tech companies will make for us. The hands-off posture — Section 230’s shield, the deference to innovation, the assumption that tech leaders would act in the public interest — has been tested at scale and failed. The warnings came from inside the companies themselves — Meta, in particular: the Haugen disclosures, the Béjar testimony, the internal documents now part of the KGM trial record. The leadership chose not to act. A generation of children paid the price.

The decision cannot be left to tech companies. The public has to make it. We have to make it.

What can be said now, in the spring of 2026, is that the work has begun. A generation of parents, teachers, and kids themselves stopped accepting that this was normal. The KGM verdict in Los Angeles cracked the legal architecture. Australia raised the age. More than thirty U.S. states have enacted or recommended classroom phone restrictions. Students started organizing phone-free spaces of their own. None of these alone closes the gap between the harm Haidt and Price named on The Daily Show in February and the product still operating today. But the work of closing it has begun.

A generation of children was harmed by design. Whether the next generation grows up under different terms is the question now being answered.

Note: Prefer to listen? Use the Article Voiceover at the top of the page, or find all narrated editions in the Listen tab at solvingfor.io.

Solving For is a deep-dive series that takes on one pressing problem at a time: what’s broken, what’s driving it, and what a path forward might look like.

Previous series have examined rare earth dominance, AI safety, the decline of local news, the end of amateurism in college sports, shrinking competition in Congress, and a world rearming as the global rules-based order weakens. Learn more at solvingfor.io.