This story was originally published in The Highlight, Vox’s member-exclusive magazine. To get access to member-exclusive stories every month, become a Vox Member today.

For more than a century, the world has run on coal.

When Thomas Edison’s Pearl Street electrical station in Lower Manhattan fired up in 1882, it ran on coal. Coal survived the oil era, the nuclear era, the dash for natural gas, and decades of back-and-forth climate policy. From the 1970s through the mid-2010s, coal supplied somewhere between 35 and 40 percent of the planet’s electricity, a steady if sooty presence powering modern life.

Then last year, it lost the lead. According to Ember’s Global Electricity Review 2026, recently released in time for Earth Day, renewable sources produced 33.8 percent of the world’s electricity last year, compared to 33 percent for coal. It was the first time those two lines had crossed since 1919, when the global grid was still small enough to run mostly on hydropower.

As coal has declined — at least on a relative basis — the sun has risen. When the Paris climate agreement was signed in 2015, solar produced just 256 terawatt hours of electricity globally. Nuclear power plants, at the time, were pumping out about 10 times that, while wind was responsible for three times as much electricity as solar. 

A decade later, solar is producing 10 times more power: 2,778 TWh, roughly what the entire European Union consumes in a year. Its production has doubled in the past three years alone. For 21 years running, solar has been the fastest-growing source of electricity on the planet. In 2025 it surpassed wind for the first time, and is now on pace to pass nuclear this year.

While the world still burns a huge amount of coal — some 8.8 billion tonnes in 2024, according to the International Energy Agency (IEA) — solar alone covered 75 percent of the rise in global electricity demand. Put wind and solar together, and you’ve met 99 percent of it. Fossil fuel power generation — coal, oil, and gas combined — fell 0.2 percent in 2025, the first decline since the pandemic and only the fifth year this century that fossil generation didn’t rise. 

Clean sources are now growing fast enough, on their own, to absorb just about everything the world is adding to its grid. And there’s a decent chance that, thanks in part to what’s happening right now in the Middle East, that transition may speed up.

Why solar is no fluke

It all starts with cost.

Solar module prices have fallen roughly 75 percent every decade for more than 40 years, a pattern so durable it has its own name, Swanson’s law, the observation that the price tends to drop by 20 percent every time the total number of solar panels ever built doubles. This rule has held through supply gluts, trade wars, and pandemics. In the mid-1970s, a solar module cost more than $100 per watt. In late 2025, one panel cost about 10 cents per watt. No other major energy source in modern history has gotten that cheaper, that fast.

The oldest objection to solar — that it goes dark when the sun goes down — is becoming obsolete because we can increasingly store the daytime electricity solar units generate. Battery costs dropped 20 percent in 2024 and another 45 percent in 2025. Global battery deployment grew 46 percent last year, to 250 gigawatt-hours. Solar plants built with enough batteries to deliver power round the clock now sell electricity in the US for around $76 per megawatt hour, cheaper than building new natural gas capacity.

The China story

The world’s long-time manufacturing powerhouse — China — has made this shift possible. Chinese factories now make around 80 percent of the world’s solar panels and an even larger share of the polysilicon, wafers, and cells that feed into them, a dominance built over two decades of state-backed investment, enormous scale, and ferocious price competition. The result is the cheapest energy technology in human history, produced at a pace the rest of the world has not matched. 

Chinese dominance has also made clean power a geopolitical story: tariffs, trade disputes, arguments in Washington and Brussels about whether to build parallel supply chains. For the climate, though, the math is simple. Cheap panels built anywhere cut emissions everywhere.

The demand side has moved too. For most of the last two decades, the global coal story has been a Chinese story. When China’s electricity demand surged, so did coal. When it slackened, so did coal. That relationship cracked in 2025: China’s fossil generation fell 0.9 percent, its first decline since 2015, even as the country’s electricity demand rose 5 percent. India’s fossil fuel generation fell as well, by 3.3 percent, while its renewables grew 24 percent year over year. In both cases, new clean energy capacity outran new demand. Ember found that renewables in China now produce more electricity than every household and service-sector business in the country, combined.

Don’t get carried away — yet

A flat year for coal is not the same as a falling one. Power-sector emissions in 2025 were still close — within a rounding error — of 2024’s levels, which set a record high. In its report, Ember calls this moment “the era of clean growth,” which should be understood as the start of real decarbonization, rather than a final state of decarbonization.

Coal’s share is shrinking — from a peak of 41 percent of global generation in 2013 to 33 percent today — but the fleet itself isn’t going away. China approved more than 40 gigawatts of new coal capacity in just the first three quarters of 2025. Thanks to growth in renewables, these plants are increasingly becoming a backup source, rather than a primary one. But those plants exist, they burn coal when they run, and they’ll burn coal for years.

Then there is the US. The Trump administration’s One Big Beautiful Bill Act ended the residential solar tax credit in December and tightened eligibility for commercial projects. Rhodium Group, a research institute, projects the law will cut US clean-capacity additions through 2035 by more than half. America is in danger of getting left behind.

That sounds bad, and in the short run it is. But policy can slow a market; it has a harder time stopping one when the economics have already shifted. BloombergNEF reported that global energy-transition investment hit a record $2.3 trillion in 2025, up 8 percent from 2024. China alone put roughly $800 billion into clean energy last year; India’s clean-energy spending climbed 15 percent to about $68 billion; the EU has been accelerating renewables spending ever since Russia’s invasion of Ukraine cut its pipeline gas. Even if Washington slows down, the rest of the world is building solar farms and battery plants as fast as the supply chains allow. The US is trying to run against a market it no longer controls.

There is, however, the AI wild card. The IEA estimates global data-center electricity use rose 17 percent in 2025, with AI-specific demand growing faster. In the US, gas is currently the biggest single source of new data-center supply. Artificial intelligence is the one uncontrolled variable that could swamp clean-power gains in the back half of this decade.

Strait talk

The last big oil shock rewrote the global energy system. After the 1973 OPEC embargo, President Jimmy Carter put solar panels on the White House, founded the Solar Energy Research Institute in Golden, Colorado, and signed the country’s first appliance efficiency standards into law. Ronald Reagan undid much of that work, but the seed technologies — photovoltaic R&D, efficiency standards, CAFE rules for cars — kept developing in the background for decades.

This time, the shock is being felt by a system where clean alternatives are already the cheapest option in most places. The US-Iran war has led to the effective closure of the Strait of Hormuz, through which roughly a quarter of seaborne oil and a fifth of global LNG normally flow. The IEA called it the largest supply disruption in the history of the global oil market.

The response has been exactly what cheap clean power makes possible. In March, global solar generation grew 14 percent year over year and wind grew 8 percent; solar alone saved European buyers some $3.5 billion in gas costs for the month. Countries that might have responded to an oil crisis in 2006 by drilling faster are instead moving up construction for solar farms, offshore wind, and grid-scale storage. Where the 1970s planted seeds that took 40 years to sprout, 2026’s shock is meeting an industry already at commercial scale.

The climate case for clean power has always rested on a simple bet: that the technologies would keep getting cheaper faster than the politics got worse. Today, solar is the fastest-growing source of electricity in the history of electricity, while coal looks to be on a terminal decline. Batteries are starting to make it a 24-hour fuel. What comes next is a question of speed — and speed, mostly, is a question of choice.

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

If you’ve ever used an online patient portal to message your doctor in the middle of the night, you won’t be surprised to learn that responding to those messages takes an increasingly big bite out of clinicians’ workdays. 

So in recent years, hospitals have begun adopting an AI tool that can draft responses for them. The tool was supposed to make a time-consuming task go more quickly and smoothly, said Philip Barrison, an MD-PhD student at the University of Michigan Medical School who studies AI in healthcare.

Instead, the tool has given doctors and nurses a new to-do list. First they have to read the AI-generated response and decide if it “is actually something that they think they would say,” Barrison said. Humans are suggestible, and looking at something and deciding whether you would have thought of it on your own is a cognitively complex task.

Even if the message looks correct, the clinician still needs to “edit it to the point where they think it’s acceptable” to send to a patient, Barrison said. The AI tool introduces a totally new set of complicated judgment calls into what used to be a relatively straightforward process. As a result, many clinicians have chosen not to use it at all.

They’re fortunate to have the choice. Buoyed by expectations of cost savings and skyrocketing productivity, companies are increasingly asking (and sometimes requiring) employees to use AI to make their work more efficient. Meta, for example, last year instructed some workers to use AI to “go 5X faster by eliminating the frictions that slow us down.” The CEO of Shopify told employees they’d need to prove they “cannot get what they want done using AI” before the company would approve new hires. Some companies are even evaluating or ranking employees based on how much they use AI tools.

Workers in some sectors have found major time savings from AI. But for others, the tools just change the work rather than making it faster. Workers might be spending less time writing patient portal messages, for example, but more time editing the releases the AI tool writes. 

At best, this mismatch between employer expectations and employee reality can be an annoyance. In other cases, however, it can result in workers being laid off for failing to meet unrealistic efficiency demands. Some critics say the overzealous adoption of AI in high-stakes settings like healthcare even puts people’s lives at risk. Now workers, unions, and experts are increasingly calling for guardrails to protect employees from inflated expectations around AI — and customers, students, patients, and the general public from mistakes that can happen when managers put AI adoption above all else.

The hidden costs of AI use 

Corporations are increasingly presenting employees with a choice: Use AI to be more productive or “you’re going to be automated out of a job,” said Aiha Nguyen, director of the labor futures program at the research organization Data & Society.

But the effects of AI on productivity aren’t as straightforward as some CEOs have claimed. In one 2025 study, software developers believed AI made them faster, but in fact they took 19 percent longer to complete tasks. (The researchers tried to repeat the experiment this year but had trouble recruiting developers who would agree to work without AI.) And in a recent survey of 5,000 white-collar workers, 40 percent of rank-and-file employees said AI saved them no time at all.

Workers across heavily AI-exposed fields point to hidden timesucks that come with using the technology. Julie, an art teacher, wrote in a response to a Vox reader survey that her school’s administrators routinely suggest using AI for lesson-planning, emails, and progress report comments. She’s tried AI-generated lesson plans, but they don’t account for the fact that kids may work through an activity at different speeds.

“First, I am checking what AI suggests, then I am editing them. Why add a step I can accomplish on my own?”

Julie, an art teacher who wrote in response to a Vox reader survey

“First, I am checking what AI suggests, then I am editing them,” she said. “Why add a step I can accomplish on my own?”

For an employee at an East Coast communications agency, an internal AI tool was supposed to speed up the process of drafting press releases and other documents about the pharmaceutical industry. 

“The goal is, I think, to be able to plug and chug into this machine and be able to turn a lot of materials around a lot quicker than we already do,” said the employee, who asked to remain anonymous for fear of career repercussions.

But when the employee tried to use it for basic research, it made too many mistakes. Double-checking its work erased any time savings. When the employee tried using it for communications with clients, its people-pleasing tendencies became a problem, as the tool put a “weird happy spin” even on messages warning of bad news.

“Part of the reason we take a human speed to turn things around is because there is so much nuance behind everything that we do,” the employee told me. “AI is just not going to be able to catch it.”

It’s not just that AI makes errors. With the advent of agentic AI, workers are increasingly being asked to edit and oversee the output of multiple AI tools, a new kind of work that can have unexpected costs. 

One recent study of 1,488 workers across industries, for example, found that excessive oversight of AI agents could lead to what the researchers called “AI brain fry,” a kind of cognitive fatigue. “Participants described a ‘buzzing’ feeling or a mental fog with difficulty focusing, slower decision-making, and headaches,” the researchers wrote in Harvard Business Review. Brain fry was also associated with an increased number of errors and an increased desire to quit one’s job. 

The researchers also found that while using one or two AI tools increased productivity, adding additional tools produced diminishing returns, and after four tools, productivity actually declined. 

What workers really want from AI

Despite such findings, companies continue to pressure employees to use AI, and to cite AI investment as a rationale for layoffs, even as companies that try to link staff reductions to AI adoption tend to struggle on the stock market.

Some workers and organizations, however, are beginning to push back. National Nurses United, the country’s largest nurses’ union, has criticized the use of AI tools in hospitals to estimate staffing needs or to recommend treatment protocols for patients.

There’s no guarantee that these tools will take into account a patient’s individual profile, including underlying medical conditions, the way human clinicians can, Cathy Kennedy, the union’s president, told me. AI is supposed to “help us do our work more efficiently, but at the end of the day, it makes it even more burdensome,” she said.

Hospitals need to evaluate, with nurses at the table, whether AI tools really work as advertised, Kennedy said. “We have to stop — we have to go back and really see if this is truly doing what it needs to do,” she said.

The same is true across industries, Barrison, the healthcare researcher, told me. “Organizations need to be prepared to say when, if they were seeking a return on investment, if they were seeking value in a technology — how do you define what that value is? And if there’s not value there anymore, how do you turn it off?”

Some workers have found ways that AI actually helps them do their work — just not the ones management expected. Julie, the art teacher, likes to use Claude to learn more about topics she’s less familiar with, like kiln-firing ceramics. 

Meanwhile, researchers have found that AI can actually reduce employee burnout, if it’s used to complete tasks employees find burdensome. “Everybody in every job has a list of things that they procrastinate on,” said Julie Bedard, a managing director and partner at Boston Consulting Group who led the AI brain fry study. “Those are the places I get, unsurprisingly, a lot of enthusiasm to try AI with.”

But employers won’t find out what those burdensome tasks are unless they listen to rank-and-file employees. “Worker standards and worker rights should continue to be at the heart of all of this,” Nguyen said, “rather than just focusing too much on the AI.” 

This story appeared in The Logoff, a daily newsletter that helps you stay informed about the Trump administration without letting political news take over your life. Subscribe here.

Welcome to The Logoff: Donald Trump is still trying to implement his tariff agenda. 

Wait, didn’t the Supreme Court strike that down? Yes — this is the Trump administration’s third try at imposing global tariffs, this time using a narrower provision about unfair trade practices. 

The Supreme Court ruled in February that Trump’s “emergency” tariffs regime was illegal. And last month, a different court also struck down Trump’s first attempt to reimpose tariffs using a different — and temporary — authority. 

What is Trump trying to do this time? Earlier this year, the administration began investigating whether US trading partners like Mexico, Canada, and the European Union import goods made with forced labor in their own trade. It has now concluded that 59 countries, plus the EU, do so — and consequently wants to impose tariffs between 10 percent and 12.5 percent on those countries. 

Under the current administration plan, the tariffs would take effect next month. Some products, such as beef, coffee, and critical minerals, would be exempted from the tariffs, according to the New York Times.

What’s the context? Forced labor, obviously, is atrocious; preventing it is a worthwhile goal. That being said, there’s no real reason to believe this effort by the Trump administration comes out of genuine conviction — the EU, for example, already has new forced labor restrictions set to take effect late next year, but is still facing new tariffs. The US also isn’t immune to issues with forced-labor imports, despite laws intended to prevent them.

Instead, this is just the latest tool his administration has landed on to do what Trump has been trying to do since he took office: Impose sweeping tariffs in service of an economically illiterate concept of the US economy. 

What else should I know? Even as Trump tries to impose new tariffs, his administration is also fighting to hold onto some of the $166 billion in revenue it earned illegally from his first round of tariffs.

And with that, it’s time to log off…

Hey readers, it’s NBA Finals time! If you’re just tuning in, my colleague Benjy Sarlin has the explainer for you here (it’s a gift link). And if you’re trying to decide which team to root for, New York magazine has a bandwagoner’s guide to the Knicks here. (I’m compelled — an underdog is always fun to root for.) 

Thanks for reading, and we’ll see you back here tomorrow!

There are many ways to bomb a college commencement speech. 

You can tell everyone you composed the talk while high on ayahuasca, like Chris Pan at Ohio State. You can deliver the entirety of your speech in the voices of your incredibly annoying cartoon characters, like Tom Kenny and Bill Fagerbakke at the University of Vermont. You can even, like my graduation speaker in 2001, admonish the graduating class for depending too much on their parents and generally being an ungrateful lot, before later being convicted of multiple counts of sexual assault and undergoing a dramatic fall from grace. (Yes, that was none other than Bill Cosby, whose convictions were later overturned.) 

But the surest way to turn your graduate audience hostile in 2026 is to refer positively to AI, as speakers ranging from former Google CEO Eric Schmidt at the University of Arizona to real estate executive Gloria Caulfield at the University of Central Florida to record label honcho Scott Borchetta at Middle Tennessee State University discovered. And that’s because AI has — not unreasonably — become the symbol of growing fears that a college degree is no longer as valuable as it once was, and that today’s college grads are uniquely screwed. (The only speaker I could find whose comments on AI were well received was The Daily Show’s Ronny Chieng at Harvard, probably because they included the line: “fuck AI, fuck AI, fuck AI.”)

In a late-2025 NBC News poll, 63 percent of voters said a college degree isn’t worth it, against just 33 percent who said it was. A Gallup poll found that the share of Americans who say college is “very important” had fallen to 35 percent in 2025, a huge drop from 75 percent in 2010. And that pessimism has real grounding. Recent graduates ages 22 to 27 had an unemployment rate of about 5.7 percent in early 2026, above the national average of 4.3 percent. Hiring has slowed to the lowest rate outside the pandemic since 2014, while entry-level postings have fallen roughly 35 percent over the past 18 months. 

So there’s no doubt that 2026 will be a rough launch for new college grads. But a rough launch doesn’t mean a rough life, and while the longer-term impact of AI is unknowable, it’s far from the worst time even in recent memory to graduate into the workforce. The data still says, for most graduates, a college degree is more than worth the investment.

The vibes out there for college grads are not good. But when the bad vibes are outpacing the actual reality, that qualifies as qualified good news. 

One of the best investments you can make

Let’s start with the number the college panic ignores. In 2025, the Federal Reserve Bank of New York asked the question “Is college still worth it?” and came back with a very specific answer: Yes — to the tune of 12.5 percent. 

That was the median return on investment in a college degree, after accounting for the cost of tuition and the amount lost by not spending those years working. College graduates in recent years have earned a median of around $80,000 a year, compared to around $47,000 a year for high school graduates. Government data in 2024 put median weekly earnings for workers with a bachelor’s degree at $1,543, compared with $930 for workers with only a high school diploma — about 66 percent more. And while it’s true that the growth of this premium has largely flattened over the past two decades, after roughly doubling between 1980 and 2000, it hasn’t disappeared. Graduating from college, even in 2026, still puts you on a better path than skipping it.

It’s telling that when you shift from the abstract idea of college to the value of individual degrees, the vibes change. Asked about their own degree, according to a 2026 Gallup poll, about 80 percent of bachelor’s graduates call it critical or important to their careers, while 71 percent say they landed a good job within six months. It’s a bit like the perennial attitude toward Congress: People hate the institution and yet tend to rate their own representatives highly. Abstract views are influenced by the deluge of content about the crisis of college, while individual views are influenced by what is actually happening to people. 

It’s the timing, not the degree

Speaking as a proud member of the college class of 2001, I can tell you that 2026 is far from the first year when it was tough to graduate into the workforce. My friends one year above me in college entered an economy that had an astoundingly low unemployment rate of 1.4 to 1.7 percent for college grads ages 25 to 34, while real hourly wages for young college graduates had grown at 3 percent a year between 1995 and 2000. My classmates assumed we were headed for the same golden outcome.

“Psych!”, as we used to say back then. By the spring of 2001, the dot-com crash was in full effect, wiping out startups and jobs. More than a few people I knew had lined up lucrative starting jobs at investment banks and consulting businesses, only to have those gigs rescinded as they were preparing to receive their diplomas. (I cleverly avoided this by never getting those offers in the first place and instead entering the thriving field of journalism.) By December 2001, in the aftermath of 9/11, the unemployment rate for college grads ages 25 to 34 had jumped to 4 percent.

The class of 2010 had it even worse — recent college grads had a 7 percent unemployment rate. But though both the classes of 2001 and 2010 experienced what economists call “recession scarring” that had lasting effects on their income, those scars largely, though not completely, faded as time passed and the economy improved. The lesson? You can’t control when you graduate college, but you can largely control whether you graduate college at all — and finishing school is likely to still benefit you over the long term.

It’s true that the class of 2026 is facing an extra layer of uncertainty: the fear that AI is eating away at the bottom rung of the career ladder before graduates can reach it. Goldman Sachs finds unemployment among 20- to 30-year-olds in tech-exposed roles is up nearly 3 percentage points since early 2025, while research from Stanford has counted a roughly 20 percent drop in employment for young software developers in highly automatable jobs. 

But every time you think the case has been made that AI is causing a jobpocalypse, new data complicates the picture. Vanguard reports that employment in highly AI-exposed occupations rose 1.7 percent between 2023 and 2025, while a Federal Reserve study this year of more than a million firms found no clear connection between adopting AI and posting fewer jobs so far. At the moment, hiring problems have more to do with a cautious, high-interest-rate economy. And employer hiring plans for the class of 2026 are actually being revised upward — not the move you make while deleting the entry level.  

“To you, the class of 2026, I say…”

None of this data means that college bet is a sure thing for everyone. Tracking by the Burning Glass Institute and Strada finds that 52 percent of graduates are underemployed a year out, and 45 percent are underemployed a decade later. A college grad who takes a first job that doesn’t require a degree is 3.5 times more likely to be underemployed 10 years on. For that group, the earnings premium over a high school grad shrinks to about 25 percent — roughly the same as a college dropout.

Outcomes are also influenced by what a graduate chooses to study: Underemployment runs under 10 percent for nursing graduates and above 65 percent for criminal justice majors. (I realize telling someone who just claimed their diploma that maybe they should have picked a different major is not exactly actionable advice.) And the financing has gotten tougher — for Gen Z, it cost 32 percent of the typical American family’s annual income to pay for one year at a state university in 2021, compared to mid-20s for Gen X in the 1990s and 15 percent for Boomers in 1975. 

But generational comparisons obscure as well. When people say college doesn’t pay like it used to, they may not realize they’re comparing against a past when a far smaller and more homogenous slice of Americans got their degree: Among 25- to 29-year-olds, the share holding a bachelor’s has roughly doubled between 1980 and 2021, from about a fifth to nearly two in five. That much larger and more varied pool of graduates skews the individual outcomes, even if the average largely holds up. 

So what would I tell the class of 2026 if someone were misguided enough to put me on the dais? Mustering my best commencement-grade metaphors, I’d tell them that, yes, they are graduating into a sea of troubles, but that they are far from the first academic sailors to make such a voyage, and that the diploma they hold is still the most oceanworthy raft they can find. (Can you tell I was an English major?) And if I were so bold as to mention AI, I’d lean more Ronny Chieng than Eric Schmidt.

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

Remember DOGE, the Elon Musk-led “government efficiency” project that spread chaos during President Donald Trump’s first few months back in office, fired tens of thousands of federal employees, and then vanished almost as abruptly as it began?

If you didn’t lose your job in one of Musk’s federal employee purges, or you aren’t one of the remaining federal civil servants who has to figure out how to do your job without many of your colleagues around, DOGE is probably little more than a memory. But the legacy of this era of arbitrary firings is still being litigated in federal court, and Justice Amy Coney Barrett just handed down some very bad news for nearly every civilian who works for the federal government.

On the surface, the Supreme Court’s decision in Margolin v. National Association of Immigration Judges, which was handed down on Tuesday, is a bit removed from Elon’s brief stint as Trump’s human resources manager. The case concerns whether federal immigration judges have a First Amendment right to give public speeches about immigration law. And the full Supreme Court decided to get rid of the case using a procedural argument that has few implications for federal employees.

But Justice Clarence Thomas, in an opinion joined by Barrett, wrote a separate opinion that would allow Trump to strip all federal civil servants of employment protections that many federal workers have enjoyed since the Chester A. Arthur administration.

While Thomas often takes extreme positions, Barrett is a relative moderate who is close to the center of the GOP-controlled Supreme Court. So, if Barrett is willing to endorse Thomas’s one neat trick to abolish civil service protections, that’s a strong sign that a majority of the Court agrees with her position.

Republican judges have long backed a legal theory known as the “unitary executive,” which holds that the president must have the power to fire high-ranking government officials who lead federal agencies. But the unitary executive has not historically been understood to eliminate employment protections for civil servants and other relatively low-ranking federal employees. 

Justice Antonin Scalia’s dissent in Morrison v. Olson (1988), which is considered something akin to a holy text to proponents of the unitary executive, referred to the president’s power to “remove executive officers” — “officers” are relatively high-ranking government workers — but it did not say that the president must be able to fire every individual postal worker or Social Security clerk.

In Margolin, however, Thomas and Barrett suggest a way to collapse this distinction between agency leaders and ordinary civil servants. Trump can simply fire all of the government officials who adjudicate civil service disputes, and then civil servants will no longer have any enforceable rights.

Barrett, in other words, appears to believe that civil service protections only exist if the president wants them to exist. And if she says so, it’s likely the Court’s majority will, too.

Why civil service protections are essential to a modern government

If you watched the Netflix show Death by Lightning, which was about the brief presidency of James A. Garfield, or if you read the book the show was based on, you got a pretty good picture of what the president’s life was like before civil service reform.

As author Candice Millard wrote, when Garfield took office, the line of job seekers hoping to secure a federal job “began to form before he even sat down to breakfast.” By the time Garfield had finished his meal, “it snaked down the front walk, out the gate, and onto Pennsylvania Avenue.” As president, Garfield was expected to meet with each of these job-seekers and sort them into jobs — often based on whether they had a politically powerful patron.

This system was inefficient, as it forced the federal government to replace much of its workforce every time the White House changed hands. It diverted a simply enormous amount of the president’s attention into low-level hiring decisions. It fostered corruption, as often the only way to secure a federal job was to do favors for a senator, congressman, or some other powerful figure who could act as the job-seeker’s patron. And it made it very difficult for the government to hire highly specialized workers.

Why would someone go to the trouble of, say, getting an economics degree and becoming an expert on federal monetary policy if they knew that their job in the Treasury Department would evaporate the minute their party lost an election?

President Arthur signed the Pendleton Civil Service Reform Act in 1883, shortly after Garfield was assassinated by a disgruntled job-seeker. It was the first of several laws which ensure that the government did not have to replace every Republican postal worker or FBI agent with a Democrat if a Republican president lost an election. 

Modern civil service laws also prohibit the federal government’s political leadership from coercing civil servants into political activity. They provide protections for whistleblowers. And they generally ensure that the government will be staffed by competent professionals who provide continuity across presidential administrations.

Federal civil service laws are primarily enforced by an agency known as the Merit Systems Protection Board (MSPB). Civil servants who believe their rights as federal employees have been violated typically must file their case in the MSPB, which gets the first crack at adjudicating these sorts of disputes.

Early in his second presidency, however, Trump took several actions that appeared designed to shut down the MSPB. He fired one of the Board’s members, depriving the MSPB of the quorum it needs to operate. He also fired Special Counsel of the United States Hampton Dellinger, an official who investigates alleged violations of civil service laws and brings cases to the MSPB, and attempted to replace Dellinger with a far-right podcaster.

Since then, Trump has taken some actions to reinvigorate the MSPB. The Board now has two members, which is the minimum it needs to operate. The podcaster withdrew from consideration to replace Dellinger after Politico reported that the podcaster said he has a “Nazi streak in me from time to time.” And Trump later assigned Dellinger’s duties to US Trade Representative Jamieson Greer.

So, while there are good reasons to believe that the MSPB is significantly diminished thanks to Trump’s actions, the Board currently has the minimum amount of personnel it needs to operate. But that was not true for the first several months of the second Trump administration, when it only had one member and thus was unable to adjudicate civil service disputes.

Barrett would let Trump abolish civil service protections by firing the MSPB’s members

The most interesting issue in the Margolin case concerns what should have happened if Trump had never appointed a second MSPB member, and thus had left the Board inoperative.

A federal appeals court, the United States Court of Appeals for the Fourth Circuit, decided Margolin in June 2025, during the period when the MSPB was defunct. That court suggested that, if the MSPB is nonfunctional, then the federal judiciary must step in and hear civil service disputes that otherwise would be heard by the MSPB — because, otherwise, federal civil service laws would cease to function. 

On Tuesday, the full Supreme Court reversed the Fourth Circuit, although it did so on narrow grounds. The full Court’s opinion in Margolin states simply that the Fourth Circuit should not have opined on what happens when the MSPB is defunct, because the plaintiffs in Margolin did not raise this issue in their briefs. 

But Thomas’s concurring opinion, which was joined by Barrett, rejects the Fourth Circuit’s argument outright. He argues that federal law says that civil servants must bring employment disputes in the MSPB, and if there is no MSPB, that means that they are simply out of luck.

Thus, as a practical matter, Trump could gain the power to fire any federal worker simply by firing one of the two current members of the MSPB. If that happened, the MSPB would cease to function, and federal civil servants would be cut off from any legal remedies, even if they were illegally fired for being Democrats.

Despite the significant implications of Barrett’s decision to join Thomas’s opinion, it isn’t particularly surprising. Last July, in McMahon v. New York (2025), the Court permitted the Trump administration to fire about half of the Department of Education’s workforce. Though the Court’s three Democrats dissented in McMahon, the Republican justices in the majority did not explain their decision; it was decided on the Court’s shadow docket, and the justices often do not explain their reasoning in those cases.

Nevertheless, McMahon was an early sign that the Court’s Republican majority does not support civil service protections, or believe that those laws should be enforced. Barrett’s decision to join Thomas’s Margolin opinion also suggests that she holds that view.

It appears, in other words, that this Supreme Court wants to tear down a consensus that was reached in 1883 — that the federal government should have a professional civil service that cannot be removed simply because the Republican Party controls the White House. Barrett’s move suggests Trump has plenty of leeway to keep firing people, even if federal law is supposed to stop him from doing so.

This story appeared in The Logoff, a daily newsletter that helps you stay informed about the Trump administration without letting political news take over your life. Subscribe here.

Welcome to The Logoff: Donald Trump has a new pick for director of national intelligence. 

Who’s the new guy? Jay Clayton, currently the US attorney for the Southern District of New York and a former chair of the Securities and Exchange Commission. He is — at least by Trump standards — an uncontroversial selection: Sen. Mark Warner, the top Democrat on the Senate Intelligence Committee, told reporters Thursday that he has “great respect” for Clayton. 

What’s the context? The director of national intelligence job is opening up because its current holder, former Rep. Tulsi Gabbard, announced last month that she would resign later in June. 

But Clayton’s selection for the permanent director of national intelligence job comes after Trump made a more concerning pick last week: Pending Clayton’s Senate confirmation, current Federal Housing Finance Agency Director Bill Pulte will take over as acting director of intelligence late next week, while remaining in his housing job.

Pulte is utterly unqualified for the role, and Trump’s decision to elevate him caused a bipartisan outcry. It also appears to have sunk, for now, a bill to reauthorize a foreign surveillance program that will expire tomorrow.

Does Clayton’s nomination fix everything? Not quite. The choice of Clayton as a permanent DNI satisfies one demand from lawmakers, but for now, Pulte is still set to take over until Clayton is confirmed by the full Senate (which won’t happen before next Friday). On Thursday, Trump confirmed he plans to keep Pulte in the acting role.

When Pulte does take over, his mandate appears to be to conduct “extensive cuts” to the Office of the Director of National Intelligence, according to Reuters. There are also serious concerns about what else he could do with the job, even during a brief tenure. 

At the housing agency he currently leads, he has taken it upon himself to play the role of self-appointed mortgage fraud czar and Trump attack dog, harassing the president’s political enemies.

And with that, it’s time to log off…

Hi readers, here’s a great story about the wonder of the natural world (or, well, as natural as Brooklyn’s Prospect Park gets). My colleague Benji Jones set out to discover a new species in the park — not as unlikely as it sounds, it turns out! — and he told Vox’s Unexplainable podcast all about it. You can listen here.

Plus, the first goal of the 2026 World Cup goes to host Mexico, playing at home in Mexico City’s Estadio Azteca.

Have a great evening, and we’ll see you back here tomorrow!

You’ve probably never heard of the term “RCP 8.5” — the highest-emission scenario used by climate scientists to project the planet’s future. But if you’ve read about climate change, you’ve seen the numbers and nightmarish outcomes it produced: 4°C of warming by 2100, sometimes 5°C, sea level rising multiple feet, parts of the planet too hot for humans.

Those numbers shaped a decade and a half of climate journalism, including a lot of my own when I covered climate change at Time magazine. I didn’t always know — and didn’t always communicate — that the scenario behind the most apocalyptic, attention-getting findings was largely an attempt to imagine how bad things could get, not a true forecast. But I wasn’t alone. RCP 8.5 was a frequent background presence in climate journalism.

Last month, though, the scientists who built that scenario formally retired it. In a paper published in Geoscientific Model Development, Detlef van Vuuren and more than 40 co-authors eliminated RCP 8.5 from the scenarios that will feed into the Intergovernmental Panel on Climate Change’s (IPCC) Seventh Assessment Report, which is due in 2029. Based on falling clean-energy costs, climate policy, and recent emissions trends, the highest-emissions pathway had become, in their words, “implausible.”

I can understand if your eyes began glazing over as soon as you read “seventh assessment report,” but this shift represents real progress and hope. It means that the apocalyptic climate change future that we’ve been describing for 15 years is officially no longer on the table. Instead, a merely bad climate future — about 2.8°C by 2100 — is now the central scientific estimate. Given how hopeless our climate future has appeared at times, that really does qualify as good news.  

Counting to 8.5

Climate models can’t tell you the future on their own, because how much the planet will warm depends in large part on what humans do. So scientists build scenarios: structured guesses about how the next century might unfold under different assumptions about energy use, growth, and climate policy.

Four such scenarios were introduced in 2011 as the standard set for the IPCC, the international body of scientists that periodically takes stock of global climate research and translates it into reports for governments worldwide. Three of the four were called “mitigation” pathways — futures where the world worked to reduce greenhouse gas emissions. One, the infamous and now obsolete RCP 8.5, was the “no-policy” baseline, a future with continued fossil fuel expansion, coal use roughly five times higher by 2100, and a global population pushing 12 billion. Think of it like Dickens’s Ghost of Christmas Future, a vision of just how bad things could get if we did nothing to change our ways. 

And just like any dystopia, RCP 8.5 guaranteed attention. Between 2011 and 2020, more than 2,000 climate impact studies used RCP 8.5 as their default future. Almost every dramatic projection of crop failure, mass displacement, killing heat, and coastline retreat that any general reader ever encountered in climate change coverage depended on it.

All of those projections were plausible enough under the numbers set by RCP 8.5, but by the mid-2010s, researchers, journalists, and even official government reports were routinely calling the scenario “business as usual,” a phrase that transformed a stress test into something that sounded like a forecast. It wasn’t, and it was never meant to be. Somewhere along the way, though, that distinction got lost.

How the worst case got walked back

The world that RCP 8.5 assumed will never arrive. Global coal use isn’t on a path to quintuple; consumption has largely plateaued after decades of growth. Instead of the global population ballooning to 12 billion people, the UN’s current median forecast projects about 10.2 billion by 2100, with other reputable forecasts putting the number even lower. (All things being equal, fewer people means less emissions.)  

At the same time, the clean energy transition moved faster than almost anyone in 2011 anticipated. The cost of solar power has fallen by about 85 percent since the RCPs were published, and annual global investment in the energy transition is now over $2 trillion. Actual global emissions have tracked far more closely to what you’d expect from a world trying to reduce them than from one doing nothing at all. By 2026, Climate Action Tracker estimated that current policies put the world on course for about 2.6 degrees of warming by 2100 — still serious, but a long way from 4 or 5.

Was RCP 8.5 ever realistic? One camp of experts, led by climate scientist Zeke Hausfather and energy modeler Glen Peters, argues that RCP 8.5 was plausible in 2011, but was taken off the table by genuine policy and technology progress. The other camp, led by Roger Pielke Jr., argues that the rate of global decarbonization has been roughly linear for decades. That would mean we didn’t actively avoid RCP 8.5; it was just never realistic to begin with. Both camps agree on what counts, though: RCP 8.5 should be gone, and the planet is still on track to warm between 2.5° and 3° by 2100. 

RCP 8.5 was as much a climate journalism story as it was a climate science one. In 2017, the writer David Wallace-Wells published “The Uninhabitable Earth” in New York magazine. It was probably the most widely read piece of climate journalism of the last decade, and it was built almost entirely on RCP 8.5 projections. 

Wallace-Wells revised his view in 2022, though there has been relatively little coverage of this year’s retirement of RCP 8.5. And researchers need to catch up: Pielke Jr. estimated that as late as early 2026, 30 new RCP 8.5 studies were coming out each day on average, generating more grist for the climate ultra-doom narrative. We’ll see whether last month’s announcement finally puts it to rest.

The future is in our hands 

But even if we’ve averted doom, there is a lot of work to do to secure a safer future.

The new “medium” climate pathway — the one that reflects current policies — estimates 2.8°C of warming on average by 2100, with the likely range running from 2.1°C to 3.7°C. That would still mean drastic declines in coral reefs and accelerated species extinction, worsening water scarcity, and further sea level rise. And while we’ve taken the worst of the worst-case scenarios off the table, we’ve run out of time to keep warming below 1.5°C, and 2°C — the upper limit that the 2015 Paris Accords sought to prevent. 

And as with anything to do with climate change, this scientific shift was quickly politicized. The day before Hausfather and his co-authors published their analysis of RCP 8.5’s retirement, President Donald Trump posted on Truth Social: “GOOD RIDDANCE!”, and described the change as proof that climate science was “WRONG! WRONG! WRONG!.” Not surprisingly, Trump is the one who is wrong here, as Carbon Brief explained in detail, but his mistake shows how easy it is to take the wrong lesson from the end of RCP 8.5. We shouldn’t fall for it.

The entire point of climate scenarios like RCP 8.5 was that there was no one certain future for climate change — only multiple possible futures. Whether or not RCP 8.5 was ever possible, the enormous advances in clean energy over the past 15 years are what made its retirement certain. Now we have new futures before us, waiting for what we do next.  

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

Just a few years ago, Ben Shapiro was the defining voice of right-wing media. His podcast sat near the top of the charts. Posts from the Daily Wire, his media company, routinely dominated the competition on Facebook. His team was even coming for Hollywood, putting out “anti-woke” comedies and an epic fantasy series that cost millions per episode.

All that feels like a distant memory now. Shapiro’s social media traffic has collapsed, as the Washington Post’s Drew Harwell recently reported; the Daily Wire has gone through multiple rounds of layoffs since 2025. The epic fantasy series flopped. Shapiro’s struggle to stay relevant is clear on his YouTube page, where you can find painfully forced videos of the pundit reacting to trending culture.

So what happened? Ryan Broderick, a longtime internet culture reporter who publishes the Garbage Day newsletter, has a succinct explanation: “The age-old problem with working at the racism factory! They eventually make a new racism that includes you,” he wrote in May.

To learn more about the Daily Wire’s decline, Today, Explained co-host Noel King spoke with Broderick about how Charlie Kirk’s murder precipitated a MAGA vibe shift that has left Shapiro out in the cold, the new media figures rising to replace him, and whether we will miss Shapiro once he’s gone. (We very likely will.)

 Below is an excerpt of their conversation, edited for length and clarity. There’s much more in the full podcast, so listen to Today, Explained wherever you get podcasts, including Apple Podcasts, Pandora, and Spotify.

Explain your “racism factory” line, please. 

It was a pithy way to describe what I think is happening to Ben Shapiro right now, which is that he’s found himself on the wrong side of a far-right vibe shift that’s happening. 

The question of “Should American conservatives support Israel?” I think, has quickly become the deciding factor in canonizing the new wave of MAGA, or even post-MAGA conservatism in America. There’s a lot of creators on one side who say we should not be involved with Israel. They say that largely for antisemitic purposes, but also because they’re xenophobic and isolationists, but they know that this is a red line that they can go across. 

Ben Shapiro cannot follow them there because he is an Orthodox Jew who supports Israel and is a fairly standard conservative, all things considered. And so this is among the many other problems that Shapiro is having right now in trying to hold his digital media empire together.

Alright, so Ben Shapiro’s on one side. As you said, he is unlikely to ever turn his back on Israel. On the other side are people who are going hard at Israel and have been since approximately, I don’t know, October 8, 2023. Who are they? Who are the players here?

The biggest one is Nick Fuentes. He is the de facto leader of this far-right splinter cell movement, the “Groypers.” He’s got a live stream that he’s on every single day, and he’s just the most vile kind of far-right personality you could imagine. But you also have more and more creators, I think, sensing this vibe shift and moving towards him. 

Candace Owens was going so far as to even claim that Charlie Kirk was killed by Mossad. You also have Tucker Carlson, Megyn Kelly — a lot of these people I would sort of put in the camp of pretty run-of-the-mill conservative commentators who understand that Trump is not popular, and so they’re trying to feel out new territory there. And then you also have “manosphere” guys like Tim Dillon who have even started to kind of go against Israel. 

It is this thing that is happening, and social media, I think, always prioritizes the newest, most taboo idea. And so this would be a new taboo that has been discovered by far-right commentators.

So in that camp of people, you have critics of Israel that run the gamut from Candace Owens, who seems kind of nutty, to Megyn Kelly, who often seems pretty straight. What do they all have in common? Is it just their criticism of Israel?

No, my read on this is that it all stems from Charlie Kirk, actually.

The MAGA movement is not one movement. It is not one ideology. The 2024 winning coalition was this weird mismatch of far-right live streamers, manosphere podcasters, neoconservatives and the TPUSA/Charlie Kirk kind of middle-of-the-road MAGA people. I think Charlie Kirk was very instrumental in holding a lot of this together, if only because it seemed like — to them at least — he was possibly a replacement for Trump. 

I’ve read into it as the MAGA movement was trying to home-grow their own version of Trump. Charlie Kirk may have been that figure. He dies, and the whole thing starts to fall apart. And I have to give, unfortunately, some credit to Nick Fuentes here, who has always hated Charlie Kirk.

So Charlie Kirk is killed, and then these alliances form and they fracture and they reform and they refracture. What events of the last, say, eight months do we place in the post-Charlie Kirk’s assassination moment?

It’s a lot of reading the tea leaves of online discourse, I would say. But you know when the movement is working and when they’re all falling in lockstep with one another.

Sydney Sweeney’s jeans would be a good example of [that], or Cracker Barrel. They’ve been able to get this talking point to surface out of their DMs and into the general consciousness. And if you look back at the months immediately after Charlie Kirk’s murder, that hasn’t really been happening the same way. They’re not really working together. They’re fighting with each other a lot, and they’re also telling on each other. 

These people are very messy. Even as we speak, Ashley St. Clair is on TikTok sharing secrets from inside the MAGA movement and going on Hasan Piker’s stream. All these guys are unfollowing each other and fighting with each other. And it’s a lot of right wingers who are super dependent on internet attention and monetizing internet attention, and they’re really, really nervous about the internet landscape the same way all digital media publishers are. I think that’s having a negative impact on the stuffiest of the digital media-era people. And Ben Shapiro is the stuffiest.

There is something else that I’ve been thinking about a lot, which is: Ben Shapiro, when he started out, he was so young, and it was like this young man that appealed to people who were much older because he was super well-spoken and he was pugnacious. 

Now he just sort of seems old. He seems like he doesn’t really know what he should be doing on TikTok. He seems like he doesn’t really know who in the culture is relevant anymore. You could make the same argument about Tucker Carlson, even though he’s surviving, but he openly seems scared of Nick Fuentes. 

Do you think that the guys that we were used to are now the old guys and they know it, and the young guys that are coming after them are worse?

I would say that Ben Shapiro from the very beginning was much better at talking to old people than talking to young people. And it seems like what he was doing was creating a digital media company that looked hip and cool to old people, who would then give him money and he would spend that money on advertising and sort of dominate Facebook and create this flywheel that allowed him to grow pretty quickly. 

A lot of the weird preoccupations the Daily Wire has had with dominating Hollywood, for instance, feel very old to me. It feels like an 80-year-old conservative’s fever dream of what the internet could be. Just very strange stuff. 

I think it’s only gotten stranger in the last year or two, because it also feels like the Trump movement has kind of moved beyond the need for someone like Ben Shapiro. In the era of DOGE and Project 2025 and ICE occupations [and] JD Vance/AI stuff, none of it feels like Ben Shapiro is really in the mix anymore.

Do you think we’re going to look back in a few years and miss Ben Shapiro for his sort of sobriety?

Yes. I think that when digital publishers on the right, in the early 2010s, began to really lean into the internet, they inadvertently connected American conservatism and by extension global conservatism with the sea changes and tides of internet discourse. And that’s always going to go towards the thing that feels the most dangerous and the most taboo, because that’s what’s most exciting on social media. 

If you have every major conservative figure in America making money directly from the internet, there’s no real incentive for them to become more moderate. They’re going to be hitting themselves in the face with hammers and smoking meth and attacking people on the street and going full white nationalist, race-science Substack nonsense. We’re already seeing this. The days of Prager University or the Daily Wire trying to do a sensible conservative’s reaction to Cardi B’s “WAP” or whatever are just not going to come back.

Caitlin Squier-Roper, 45, recently discovered an intriguing club on Instagram: Philly Cooks a Book, a monthly meetup where locals prepare and share an assigned recipe from a specified cookbook. She could’ve enrolled through the group’s social media and shown up to a meeting, dish in hand, not knowing a single soul. So she held off on joining.

It wasn’t until Squier-Roper and her husband Anthony Fernandez, 42, attended the Philadelphia Activities Fair that she decided to get involved. Squier-Roper and Fernandez recently moved to Philadelphia after living in Seattle for over a decade and didn’t have a network in their new city beyond their families. When they heard about the Activities Fair, a one-day exhibition of clubs, civic groups, and community organizations enrolling new members, the couple thought it the perfect opportunity to spread their social wings. 

Thousands of other people had the same idea. 

On a Sunday in April, around 2,300 attendees crowded every inch of available space in a historic downtown civic center to discover, and potentially sign up, for a club. Outside, it was the perfect kind of spring day: abundant sunshine, a light breeze, giving way for the serendipitous pop-ins from curious passersby. Inside, spectators shuffled, shoulder to shoulder, in single-file lines up and down the building’s winding staircase and through two rooms of tables representing more than 40 clubs, including a community for Black artists, a book club but for podcasts, and an a cappella group, stopping to chat with organization leaders and join their ranks. It was in one of these glacial plods around the ground floor of the event space when I met Squier-Roper and Fernandez. They’d already signed up for the cookbook club, the a cappella choir, and a cycling group.

The event itself, structured as it was, was novel for the couple and Squier-Roper said she was nervous to attend. “It seems out of the box and vulnerable,” she told me. But, looking around the room, she was in good company. “It’s helpful to see how many other people are here in the same searching situation,” she said. “It’s pretty cool.”

If Squier-Roper and Fernandez have felt socially adrift as of late, they certainly are not alone. The 2025 American Psychological Association’s Stress in America survey found that about half of US adults reported feeling isolated, left out, or lacking companionship at least some of the time. According to Harvard Graduate School of Education’s Making Caring Common Project’s 2024 survey, 21 percent of respondents said they were seriously lonely, over two thirds of whom felt like they lacked belonging in meaningful groups. 

Loneliness has become something of a buzzword: The US surgeon general and the World Health Organization have issued warnings about its harms, and brands and startups shill their products as the potential solution. Despite the shallowness of viral marketing campaigns and AI chatbots designed to absorb the role of friends, the problem is serious. Decades of research supports the dangers of chronic loneliness and social isolation: increased cardiovascular health risks; links to personality disorders, suicide, cognitive decline, and depressive symptoms; even a higher likelihood of mortality. 

Although many Americans say they’re lonely, and perhaps have become more aware of its negative impacts, they don’t seem to be prioritizing activities that foster connection. According to the American Time Use Survey, people spent nearly half of their waking time — more than  six-and-a-half-hours — alone in 2024, compared to just under five hours in 2003. Young people spent 45 percent more time alone in 2023 than they did in 2010. What are we doing with all this time in solitude? Watching TV, staring at our phones, gaming, mostly.

Against this backdrop, a crop of community-minded organizers stumbled into a similar train of thought: People are disconnected (perhaps I am one of these people). My city has a treasure trove of hobby clubs and civic organizations. If I lead a horse to water, can I get it to drink? From this seed of an idea, a genre of connectedness events was born — the activity fair, stuff to do fair, joining fair. From Philadelphia to Oakland, a wave of well-attended one-day activity fairs are the latest grassroots efforts to combat loneliness and connect people to their communities. These festivals operate under a simple premise: getting people in a room with club representatives is more effective and less overwhelming than scouring the internet, and it lowers the barrier to entry. 

“If there are things to join, people will join them,” Pete Davis, co-director of the documentary Join or Die, told me. But first they have to find them.

A nation of clubs

From the dawn of civilization, humans have hung out in group settings. The Romans had professional organizations known as collegia, medieval Europe had guilds, Victorian England had (exclusively male) social clubs. In the United States, people formed and joined groups of all kinds, from the Freemasons and abolitionist societies to women’s suffrage clubs and the Elks.

But participation in these groups has declined, as political scientist Robert Putnam famously explained in his book Bowling Alone: The Collapse and Revival of American Community, first published in 2000. Putnam found that enrollment in clubs of all kinds had dropped since the 1960s. Conditions have seemingly not improved. The inaugural Social Connection in America report, released last year, found that two-thirds of participants don’t belong to or never attend a meeting of any sort of organization or club. A 2024 survey from the Survey Center on American Life found that fewer than two in 10 Americans were members of hobby or activity groups, neighborhood associations, sports leagues, or parent groups.

Group membership confers many benefits. Research has shown that joining a community group led to reduced loneliness and increased social support for older adults. A scientific review found that sports team participation improved well-being, reduced stress, and increased social functioning. Being in multiple groups makes people happier. 

The regularity with which you meet makes clubs effective friendship-builders: If you see someone frequently enough, it’s easier to forge a relationship with them. Even if full-fledged friendship isn’t the goal, simply making acquaintances is sufficient to stave off loneliness and foster a sense of belonging. As Putnam wrote in Bowling Alone, “As a rough rule of thumb, if you belong to no groups but decide to join one, you cut your risk of dying over the next year in half.”

America has appeared to be club-curious as of late. In the years following the Covid-19 lockdowns, many people have yearned for tangible social connection, with the proliferation of supper clubs, run clubs, silent book clubs, and other activity- and identity-focused groups. “What I’m seeing is really in this last year, such a renewed interest in hobbies, hobbies for health,” said Julia Hotz, the author of The Connection Cure: The Prescriptive Power of Movement, Nature, Art, Service, and Belonging.

Participation “on-ramps”

Activity fairs are the natural next step in bridging the gap between the crop of niche and hyper-local clubs and a curious, but overwhelmed, populace. The concept is no different than welcome week activities at college campuses where a quorum of university clubs table and recruit new members. Designed to reach members of the wider community, club fairs operate as a live directory of a city’s offerings, all under one roof. “There’s no shortage of Instagram accounts and apps of things to do, concerts, events,” Brian Adoff, the founder of Join Philly, the group that organized the Philadelphia Activities Fair, told me. “And people still can’t find stuff.”

Adoff has long understood the benefits of clubs. In 2023, he and a friend founded a choir, bringing strangers of all ages to bars for impromptu concerts. Many attendees, he noticed, were attending solo, new to the city, or both, and formed friendships from the group. But it wasn’t until he attended a screening of Join or Die, a 2023 documentary extolling the benefits of joining clubs, compounding on Putnam’s work, that Adoff thought, I want to do this.

When I checked in with Adoff at the Philadelphia Activities Fair, he was standing on a stage overlooking the ground floor of the event, getting a good glimpse of the hundreds of locals learning about the dozens of clubs he’d brought there. 

Join Philly initially began as an online directory of clubs and associations, and the issue wasn’t finding clubs to showcase — there were plenty of those. It was getting people to participate. Sure, locals could scour the internet for a hobby group, but what if they weren’t even sure what to search for? What if they’re a little shy and walking into a room full of people they don’t know makes their stomach turn? Putting the club-curious in the same room as the groups solved some of these issues, a concept Adoff refers to as a participatory “on-ramp.” “That was the first on-ramp,” Adoff said. “How do we make this easier?”

C.C. Tellez, 48, the executive director of Lez Run, an LGBTQ+ running club, found this direct approach effective at quelling prospective members’ concerns. “Online, people like the idea of something, but they’re afraid to take the first step,” Tellez told me over the thrum of the Philadelphia Activities Fair. Tellez was approached in person by people who follow Lez Run on Instagram but were concerned about the pace, about being new. “We let them know we welcome everybody: different paces, different setups, whatever you’ve got going on, we welcome it here,” Tellez said.

“Versions of you”

I first became aware of activity fairs in 2025, when I learned of one happening in Lancaster, a small Pennsylvania city not far from Philadelphia. Hundreds of people crowded into a community center in the middle of winter to learn about a brewing club, rugby team, a mechanical keyboard club. 

The event’s organizer, Sav Thorpe Capizzi, had a lightbulb moment after a friend invited her square dancing a few years ago, something she’d never done before. As she do-si-do-ed with strangers, Capizzi wasn’t worried about how she looked or her skill (or lack thereof). “I just felt so alive in that moment and I was just so grateful to the part of me that just said yes to potentially looking a little foolish,” she told me. “And I was like, ‘Okay, so I need to invite everybody I know to every club I can think of right now.’”

Unaware of any other event geared specifically toward adults, Capizzi sent an email first to the town library, then a guild of crafters, and eventually cobbled together a list of exhibitors. She dubbed her version the Stuff to Do Fair and from that initial event sprung an offshoot in a smaller Pennsylvania town and, also the more robust second-annual Lancaster Stuff to Do Fair. This year, Capizzi doubled the amount of exhibiting clubs to 50 and nearly 600 people attended, she said.

In Capizzi’s estimation, fear of being bad at something is the biggest barrier to entry for any potential new club attendee — feeling like you’re not the kind of person who has the body for roller derby or the wit for improv comedy. Activity fairs give the shy, the uncertain, the hesitant permission to imagine themselves as someone who does. “There’s this air of novelty of all of the versions of you that exist at each of these tables,” she said. “It’s exciting, it’s thrilling, and I think it really gives people the opportunity to see themselves actually becoming the kind of person who enrolls in a class or takes up sketching after so many years.”

An excuse to be social

The Stuff to Do Fair, as well as the Philadelphia Activities Fair and the other club fairs I came across in my reporting, were organized by individuals, and perhaps that’s part of their charm. They’re scrappy and community-driven. But it’s easy to imagine a world in which these events might be sanctioned by local governments to promote public health. Social prescribing, a practice where patients receive a script not for pills but attendance at a community group, has gained momentum around the world, with medical professionals connecting patients to cycling clubs, performing arts groups, or volunteer organizations. Recent research has found social prescribing in the United Kingdom, where it was first developed, has led to improvements in wellbeing, happiness, and life satisfaction.

In place of a medical professional linking individuals to groups and activities that might benefit their mental or physical health, there could be activity fairs. “In other countries where we have more government support for social prescribing programs, what that government support goes to are up-to-date databases of the different activities that exist,” Hotz, the author of The Connection Cure, told me. If online listings and databases are out of date, well-intentioned would-be participants could be easily deterred, however motivated they might be. Solely relying on the internet to disseminate cub information means those with unreliable access or who aren’t tech-savvy are shut out from opportunities, too.

“An activity fair, giving you the information in real time and letting you meet with the people part of it in real time, I think just goes such a long way in making sure that your interest becomes a reality,” Hotz said.

Social prescribing gives people permission to do something meaningful, and to be convivial in the process. And so do activity fairs. “What a joining fair is is answering the question, What are you doing alone that you could be doing together? by having a cooperative recruiting event,” said Pete Davis, one of the directors of Join or Die.

In addition to a traditional screening tour, Davis, and his co-director and sister Rebecca (who was a supervising producer for the second season of Vox’s Netflix show, Explained) helped dozens of community organizers across the country host their own joining fairs in order to promote their film. But even if event planners didn’t work with the Davises directly, their documentary served as a point of inspiration.

Like Adoff in Philadelphia, Jared Joiner watched the Davises’ documentary, and it set the wheels in motion for his own fair in Oakland. That his actual last name is Joiner is not lost on him. “I had not thought about it in this world of joining clubs and joining organizations until the first time that I watched Join or Die and they say ‘joiner’ so many times in it and I was like, ‘Oh my gosh, am I destined to do this work?’” he told me. 

The same day that thousands of Philadelphians signed up for clubs at Join Philly, Oakland hosted its first Join-Up at a brewery in the midst of torrential downpours. Although the event had a smaller footprint — about 250 attendees and 22 organizations — many club representatives told Joiner they ran out of sign-up sheets. 

Catalysts for connection

There’s something to be said about the kind of person who attends an activity fair. “There’s definitely a self-selecting group that’s like, ‘I’m going to get off the couch to go to this thing, so I better sign up for stuff once I’m there,’” Joiner says. But there are a myriad of motivating factors getting those people off the couch in the first place: a recent move, a loss, a birth, a new job, a desire to unplug, to learn a new skill. 

The consequence, deliberate or not, is the forging of new social connections. When our worlds shrink to the confines of our homes and our screens, intentionally exposing ourselves to newfound (or newly rediscovered) activities and new people helps broaden them again. Clubs, with their regular, predetermined cadence and specific focus, are the ideal entry points to connection: Striking up a conversation is simple when you already have something in common — the activity itself. 

Toward the end of my afternoon at the Philadelphia Activities Fair, I ran into Deborah Winter and Terry Borden, both 71, as they finished a lap on the first floor of the exhibition. Winter is moving to Philly soon, and although Borden has been a resident for two decades, they both are still on that never-ending path toward community. “It’s hard to find your people,” Borden told me. 

Through clubs they signed up for — a book club, a skill-share — they hope to find both friendships and more casual relationships, something they’re already practicing. As it turns out, the two women are new friends themselves, recently introduced by a mutual. 

Even if they didn’t branch out at any of their new groups, I thought, at least they could reminisce, some day in the future, about this event, about the time they went out on a limb and joined a few clubs. Maybe the groups were boring or not quite the right fit, maybe they weren’t. But they tried something unfamiliar, together, and that’s something.

As the world waits to see if President Donald Trump will give his final approval to a deal to reopen the Strait of Hormuz and, perhaps, finally bring the 2026 US-Iran conflict to a close, it’s already clear that one of the more surprising developments of the conflict has been the prominent role of Pakistan as a mediator. 

It was Pakistan’s military leader, Field Marshal Asim Munir, who served as the key go-between in the talks that led to the initial two-week US-Iran ceasefire in early April, and Prime Minister Shehbaz Sharif who announced that it had gone into effect. Several days later, Islamabad hosted the highest level talks between the US and Iranian governments since 1979, including US Vice President JD Vance. On April 21, Trump announced the ceasefire had been extended, saying it was at Pakistan’s request. Munir has made two personal visits to Iran as part of his mediation efforts, the most recent on May 21.

Whereas the “P5+1” countries of the UN Security Council — the US, China, the UK, France, and Russia, plus Germany — helped bring about the 2015 Iran nuclear deal, and Oman hosted the US-Iran talks in the lead-up to the war, Pakistan has been the intermediary and negotiating venue of choice since the conflict began. The world’s only predominantly Muslim nuclear power is a rare country with credibility on both sides of this war. 

Pakistan’s prominent diplomatic role in the conflict is the latest sign of the unexpectedly close relations between the country’s government and the second Trump administration. “Thank you to Pakistan and its great prime minister and field marshal, two fantastic people!” Trump wrote in a characteristic Truth Social post in April. He has lavished particular praise on Munir, whom he has called an “exceptional man” and “my favorite field marshal.”

Pakistan’s new role as an indispensable US partner is partly due to some skilled Trumpian diplomacy by its government and partly due to just how much this administration’s global priorities have changed from the days when China and jihadist terrorism were the top of the agenda. 

How Pakistan went from pariah to partner in Washington

All of this would have been difficult to imagine during Trump’s first term, when Pakistan was often treated as a pariah. 

On New Year’s Day in 2018, Trump suspended most security assistance to Pakistan, tweeting, “The United States has foolishly given Pakistan more than 33 billion dollars in aid over the last 15 years and they have given us nothing but lies & deceit, thinking of our leaders as fools.” 

Trump would go on to cancel hundreds of millions of dollars in aid to Pakistan, which had been a close US counterterrorism partner even amid widespread allegations that it had provided safe harbor to the Taliban militants fighting US forces in Afghanistan and maintained relations with other anti-US militants. Pakistan responded by halting intelligence-sharing with the US amid widespread anti-American protests.   

At the same time, Trump cultivated a close relationship with Pakistan’s arch-rival India and its prime minister, Narendra Modi. Modi’s brand of majoritarian populist politics made him a natural Trump ally, and India’s position as a superpower counterweight to China made it a natural security partner for the US. The pro-Indian tilt in US foreign policy continued into the Biden administration, and there was every expectation it would carry through when Trump returned in 2025. 

Flattery and crypto: How Munir won over Trump

Pakistan’s turnaround with the new Trump administration began in early March 2025, when the country arrested an ISIS-K operative who was allegedly a key planner of the Kabul airport suicide bombing that killed 13 US troops during the withdrawal from Afghanistan, and extradited him to the United States, earning public gratitude from Trump.  

Then came the brief May 2025 war between India and Pakistan. Pakistan’s government publicly praised Trump for his “pivotal leadership” in the diplomacy that ended the conflict and nominated him for a Nobel Peace Prize. The flattery worked: Trump brought up Pakistan’s nomination during a phone call with Modi and was reportedly irritated that the Indian leader did not follow suit and, by contrast, seemed to go out of his way to downplay America’s role. 

Pakistan has also seemed particularly well-attuned to the personalist style of diplomacy in the Trump era, where the line between business and politics can be extremely blurry. Pakistan’s finance minister has signed a deal with World Liberty Financial, the cryptocurrency company co-founded by Trump’s sons and the sons of his diplomatic envoy, Steve Witkoff. 

Last year, Sharif also signed several memoranda on deals to deliver critical minerals and rare earth elements from Pakistan to the US. Pakistani officials have taken to referring to counterterrorism, critical minerals, and crypto as the “3 Cs” underlying their relationship with the Trump administration. 

The current relationship has also doubtless been helped by the ascendance of Munir, a man who Trump might describe as a military strongman out of “central casting.” Pakistan would certainly not be playing the same role today if Imran Khan, the former cricket star turned anti-American populist prime minister — who took power halfway through Trump’s first term — were still in office. Khan was removed in a vote of no confidence in 2022, which Khan blamed on the military establishment, and has been detained on corruption charges since 2023. With his removal, the military moved quickly to consolidate power. 

Pakistan’s military has always played a significant and complex role in Pakistan politics, exercising a significant amount of power behind the scenes; the country has suffered several military coups. Since Munir, formerly chief of the country’s powerful military intelligence agency, was appointed army chief by Sharif in 2022, the nation has veered closer to an outright military dictatorship: A constitutional amendment passed in 2025 gave Munir full control over all branches of the military including the nuclear forces, for the duration of a term that could last until 2030, and immunity from prosecution.  

Trump has helped cement Munir’s status by hosting the field marshal for a working lunch at the White House — the first time a Pakistani military leader rather than its elected prime minister has been hosted for such an event. 

How Pakistan is navigating America’s new priorities

If things are different now for the US and Pakistan, it’s partly just because the world is different. The US military’s withdrawal from Afghanistan in 2021 removed one of the major sources of tension in the US-Pakistan relationship: the Pakistani government’s alleged double game with the Taliban. In fact, Pakistan and the now Taliban-controlled Afghanistan have been fighting a brutal border conflict for months.

It also helps that the Trump administration is generally less focused on Islamist terrorism this time around. It has pivoted away from “great power competition” with China, decreasing the importance of India’s role. US-India relations are generally frostier over a variety of issues ranging from India’s agricultural protectionism, to immigration in the US, to India’s economic relationship with Russia. 

“The second Trump administration, in its foreign policy, is aggressively transactional; it’s not changed by strategic considerations, even compared to how it was during its first term,” said Michael Kugelman, senior fellow for South Asia at the Atlantic Council. “So in that regard, [the Trump administration] would not have any concerns about embracing Pakistan, even though Islamabad has a very close alliance with Beijing.”

Pakistan has been accumulating an unlikely set of friends and partners in recent years. Even amid its rapprochement with the US, Pakistan has deepened its military and economic relationship with China. (Xi Jinping hailed his country’s “unbreakable” friendship with Pakistan during a visit by Sharif last month.)

In 2025, Pakistan signed a nuclear defense pact with Saudi Arabia. This is particularly notable given Pakistan’s possession of nuclear weapons: Some analysts saw this as effectively extending Pakistan’s nuclear umbrella to its allies in the Persian Gulf, though others disputed this interpretation.

Pakistan’s relations with Saudi Arabia’s rival, Iran, are complex, to put it mildly. It was only in 2024 that the two countries were lobbing missiles at each other’s territory, but they quickly deescalated the tensions; they have since cooperated in combating separatist militants and smugglers along their shared border. Munir, in particular, is believed to be deeply familiar with Iran’s military establishment from his days as Pakistan’s spy chief. 

“They have proven remarkably adept and agile in ensuring that they’re able to keep all of these balls in the air,” said Elizabeth Threlkeld, director of the South Asia program at the Stimson Center, referring to Pakistan’s global web of alliances. “But they are also vulnerable to a number of different shocks from different sources, given their positioning right now.”

Pakistan’s involvement in US-Iran diplomacy is not just an effort to gain favor with Trump. Islamabad genuinely needs the war to be over as quickly as possible. Pakistan is one of the countries most exposed to the economic impact of the war: It normally imports almost two-thirds of its natural gas and 30 to 40 percent of its total imports via the Strait of Hormuz. Food and fuel prices are surging in the country. Add to that the strong domestic opposition to the US-led war among Pakistan’s population, particularly its large Shiite minority. Pakistan’s defense pact with Saudi Arabia also raises the risk of it being drawn into a conflict in the Gulf. 

If the war has highlighted Pakistan’s diplomatic savvy, it has also at times exposed its limits. For all its efforts, Pakistan’s mediation has been unable to turn April’s ceasefire into a permanent end to the conflict that reopens the Strait. At times, Pakistan has appeared to be misrepresenting the sides’ actual positions in hopes of pushing a deal through. Trump’s recent demand that a number of Muslim countries including Pakistan join the Abraham Accords as part of a final Iran deal did not go over well in Pakistan, which has refused to recognize Israel since its founding. 

The longer the war goes on, the more Pakistan’s involvement will look less like a diplomatic masterstroke and more like a credibility-taxing quagmire. As India’s experience has illustrated, foreign governments are often lavished with praise by Trump only so long as they’re useful. If Pakistan can’t deliver the ceasefire deal Trump is looking for, or if his priorities simply shift again, it may once again find itself on the receiving end of Trump’s attacks. 

This story appeared in The Logoff, a daily newsletter that helps you stay informed about the Trump administration without letting political news take over your life. Subscribe here.

Welcome to The Logoff: Last week, we wrote about President Donald Trump’s plan to create a nearly $2 billion slush fund using taxpayer money. Today, he made it happen. 

What’s the latest? On Monday, the Justice Department announced a $1.776 billion “Anti-Weaponization Fund” as part of a settlement with Trump, who previously sued the IRS for $10 billion over the leak of his tax returns. 

The fund is intended to compensate “victims of lawfare and weaponization,” a group that could include people prosecuted in connection with the January 6 Capitol riot.

According to the DOJ, “the Fund will consist of five members appointed by” acting Attorney General Todd Blanche, who is also Trump’s former personal lawyer. Trump will be able to dismiss any of those five members at will.

What else did the government agree to? The DOJ said in a press release that, in addition to ending his IRS suit, Trump would drop two administrative claims demanding $230 million from the government over the 2016 investigation into his presidential campaign and the later FBI raid of his Mar-a-Lago property. 

What else Trump gets out of the agreement is unclear, but there’s almost certainly more — the DOJ website stresses that the new “Fund” is only “a part of the settlement agreement.” 

What’s the context? Trump and his allies have already profited massively off his second term, but his administration’s corruption is reaching new heights. Another example from this week: his very active interest in stock trading. 

What’s the big picture? As Tad DeHaven wrote for Vox earlier this year, Trump has long pushed “to create discretionary pools of money and leverage points of control that can be used to reward, punish, and command, all while trying to dodge legal and constitutional constraints.”

This fund is an expression of that impulse: a vast trove of taxpayer money he can dole out to allies as he sees fit, overseen by an (acting) attorney general personally loyal to him.

And with that, it’s time to log off…

One of the goals of this newsletter is to be respectful of your time and brain space when it feels like there’s too much news to keep track of. Hopefully we’re succeeding at that — but if you still feel like there aren’t enough hours in a day, you can turn to my colleague Bryan Walsh, who wrote about how to excavate free time you might not even realize you had. (As always, it’s a gift link.) 

Have a good evening, and we’ll see you back here tomorrow!

This story originally appeared in The Guardian and is republished here as part of the Climate Desk collaboration.

The Trump administration’s plan to dismantle an ocean observation system vital to understanding the climate crisis and marine ecosystems would “severely degrade” the accuracy of weather predictions and El Niño forecasts, with economic consequences for the United States, European and American scientists have warned.

Decommissioning the US system, which plays a major part in a global ocean observation network, would lead to a massive increase in error in the annual estimates of ocean heating rates, according to research published last month.

As a result, the forecasts and early warning systems for storms, tropical cyclones and El Niño would degrade, “sometimes dangerously so,” according to Sabrina Speich, an expert in global ocean monitoring at the Ecole Normale Supérieure (ENS) in Paris and chair of the ocean expert panel of the Global Climate Observing System.

The Ocean Observatories Initiative (OOI), run by the US National Science Foundation, is a vast network of seafloor systems, underwater gliders and moored surface platforms that feeds data to researchers, policymakers, educators and mariners worldwide. The initiative, which covers both US coastlines and extends into the North Atlantic and Southern Ocean, has been used to study marine heatwaves, harmful algal blooms, subduction zone earthquakes, ocean acidification and fisheries variability.

Dismantling it would remove a major component of the Global Ocean Observing System (GOOS), a network of robotic floats, moored buoys and research vessels experts describe as the “eyes and ears” of the ocean. The warning systems based on the data, “save lives,” experts say.

Prescient research published in Nature Climate Change last month showed how data losses in GOOS, a UN-coordinated framework for ocean data for weather and climate collected by several countries, could degrade the ocean heat estimates that underpin weather prediction, El Niño forecasting and fisheries management. Losing US observations would be worse than randomly losing 80 percent of all ocean data worldwide, it found. US-funded platforms span every ocean basin, plugging critical gaps that no other nation currently fills.

“Ocean heat content is the most robust indicator of climate change we have — not just of what is happening in the ocean, but of the entire climate system,” Speich, a co-author of the research, said. Vertical temperature profiles that provide ocean heat content, are among “the simplest measurements we can make,” she said.

“Lose them, and you lose your ability to track not just ocean warming but the climate system as a whole — they are a proxy for variables that become unavailable the moment the observations stop,” Speich said. “Forecasts would continue — but they would degrade, sometimes dangerously so. Atmospheric observations alone are not sufficient. Ocean data [is] fundamental to early warning systems for tropical storms, cyclones and El Niño. And the consequences would not stop at science: the economic costs would be felt within the United States itself, from agriculture to insurance to disaster response.”

The loss of US observations, in a year predicted to be an El Niño year, with “supercharged” weather extremes, could also “lose the ability to see it coming clearly to act in time,” she said.

“The stakes are concrete: Farmers in the US and across South America use El Niño forecasts to decide what to plant and when — whether to expect drought or flooding shapes every agricultural decision months in advance,” Speich said.

The most recent El Niño, which hit in 2023–2024, was one of the five strongest on record and contributed to 2024’s record-breaking increase in global temperature.

Removing US observations alone would produce a 163 percent increase in error for annual ocean heating rates, the research by Speich and her co-authors found.

On Thursday, the European Union said it would boost its own monitoring of the world’s oceans by investing in a $107 million initiative called OceanEye, more than half of which will go to GOOS. The announcement, by the European Commission, was long-planned, not a direct response to the US move.

John P. Abraham, professor of engineering at the University of St. Thomas, in Minnesota, and co-author of the research paper, described the US administration’s move to dismantle the $368 OOI system as “penny-wise, pound foolish.”

“The US government wants to save less than a billion in sensors, which are the eyes and ears of the ocean,” Abraham said. “We have hundreds of billions in climate costs per year. The cost of the observation system is a fraction of the climate costs from hurricanes and storms that hit the US.”

The US suffered more than 400 climate and weather disasters where damages exceeded or reached $1 billion, between 1980 and 2024. In 2024 alone, the costs of such disasters amounted to $177 billion. This “billion-dollar climate and weather product,” managed by the National Oceanic and Atmospheric Administration (NOAA) will no longer be updated due to “evolving priorities,” according to a note on its website.

The system, is, Abraham said is “quite an inexpensive way to reduce climate-related costs.”

“This is not about saving money, this is about killing climate science research,” Abraham said.

Samantha Burgess, the strategic climate lead at the Copernicus Climate Change Service (C3S), the European Union’s Earth observation system which integrates European space data with in situ measurements to monitor changes and provide forecasts, said ocean observations are “irreplaceable” because “we can’t see the deep ocean from space.” They “save lives” by warning us of severe storms, she said.

“We need international cooperation to get the best available observations to mitigate risks in our changing world. Without ocean observations we are flying blind,” Burgess said.

A statement earlier this week by the National Science Foundation, which funds and oversees the OOI, said the program was not being cancelled entirely and described the plans as a “descope,” or reduction of elements, though it was not clear what data collection capacity would be left.