This is an edition of The Atlantic Daily, a newsletter that guides you through the biggest stories of the day, helps you discover new ideas, and recommends the best in culture. Sign up for it here.

The critics and the gamers have written much about The Last of Us, the video game that became a majestic HBO series. The main story is about love and family, but there’s a dark and nagging question in the scenario: If the world had no more rules, what kind of person would you be?

First, here are three new stories from The Atlantic:

The four quadrants of American politics What have humans just unleashed? You can’t define woke.

Who Are You?

This story contains spoilers for the entire first season of The Last of Us.

Did you read that disclaimer? No, I mean it—I am going to spoil everything in the first season. You’ve been warned.

In interviews, the writers of The Last of Us have said that they intended the series to be about love. And they have indeed created a gorgeous—and disturbing—tale of how we find and cherish family. But I want to raise another question that lurks in the adventures of Joel and Ellie, a dark rumble of a thought that most of us would rather not confront: If the world ended, and all of the rules of society vanished, what kind of person would you be?

This question, I think, resonates more with us today than it did during the Cold War. Back then, and particularly in the 1970s and ’80s, postapocalyptic fiction included an entire pulpy genre that the scholar Paul Brians called “Radioactive Rambos,” in which men—almost always men, with a few notable exceptions—would wander the wasteland, killing mutants and stray Communists. (They also had a lot of sex.) Sometimes, these heroes were part of paramilitary groups, but most typically, they were the classic lone wolf: super-skilled death machines whose goal was to get from Point A to Point B while shooting everything in between and saving a girl, or a town, or even the world.

But we live in more ambiguous times. We’re not fighting the Soviet Union. We don’t trust institutions, or one another, as much as we did 40 or 50 years ago. Perhaps we don’t even trust ourselves. We live in a time when lawlessness, whether in the streets or the White House, seems mostly to go unpunished. For decades, we have retreated from our fellow citizens and our social organizations into our own homes, and since COVID began, we’ve learned to virtualize our lives, holding meetings on glowing screens and having our food and other goods dropped at our doors by people we never have to meet.

We also face any number of demagogues who seem almost eager for our institutions to fail so that they can repopulate them in their own image and likeness.

Living in a world of trees and water and buildings and cars, we can posture all day long about how we would take our personal virtues with us through the gates of Armageddon. But considering that we can barely muster enough civic energy to get off our duffs and go vote every few years, how certain are we about our own bravery and rectitude?

Although Joel and Ellie are rendered with wonderful complexity by the show’s writers and by the actors Pedro Pascal and Bella Ramsey, some of the greatest moments in The Last of Us are with people the protagonists encounter during their travels: Bill, the survivalist (played by Nick Offerman in what should be a slam-dunk Emmy nomination); Kathleen, the militia leader (Melanie Lynskey); and David, the religious preacher and secret cannibal, played with terrifying subtlety by Scott Shepherd. (I warned you there were spoilers.)

Each of these characters is a challenge, and a reproof, to any of us who think we’d be swell folks, and maybe even heroes, after the collapse of civilization.

Bill is a paranoid survivalist who falls in love with a wanderer named Frank. They live together for years and choose suicide when Frank becomes mortally ill. It’s a marvelous and heartbreaking story, but Bill admits in his suicide note that he always hated humanity and was initially glad to see everyone die. He no longer feels that way, he says, implying that Frank’s love saved him, but right to the end, he remains hostile to almost everyone else in the world—just as he was before Outbreak Day.

Kathleen leads a rebellion in Kansas City against FEDRA, the repressive military government that takes over America after the pandemic. Her “resistance,” however, is a brutal, ragtag militia, and Kathleen is a vicious dictator who is no better (and perhaps worse) than the regime she helped overthrow. She promises clemency to a group of FEDRA collaborators, for example, and then orders them all to be shot anyway. “When you’re done, burn the bodies,” she says casually. “It’s faster.” She even imprisons her own doctor, who pleads with her, “Kathleen, I delivered you.” She executes him herself.

What’s important about Kathleen, however, is that she later admits that she really hasn’t changed. Her brother was the original head of the resistance: kind, forgiving, a true leader. She admits that she never had that kind of goodness in her, not even as a child—which raises the troubling thought that we all live near a Kathleen who is tenuously bound only by the restrictions of law and custom.

And then there’s David.

History is replete with times when desperate human beings have resorted to cannibalism, and although we recoil in disgust, we know it can happen. David hates what he felt he had to do, and he admits his shame. But it turns out that what makes David evil is not that he eats people but that he’s a fraud: He cares nothing about religion; he cares about being in charge, and he admits that he has struggled all his life with violent impulses. He is another character whom the apocalypse reveals more than it changes. When he gleefully tries to rape Ellie, she kills the former math teacher in self-defense.

Again, this raises the creepy question of how many Davids walk among us, smiling and toting algebra books, restrained from their hellish impulses only by the daily balm of street lights and neighbors and manicured lawns. We should be grateful for every day that we don’t have to know the answer.

Related:

The Last of Us makes the apocalypse feel new again. How The Last of Us cherishes a bygone world

Today’s News

Turkish President Tayyip Erdoğan endorsed Finland’s NATO bid; he has not yet approved Sweden’s. The Justice Department is reportedly investigating the surveillance of Americans by the Chinese company that owns TikTok. President Joe Biden urged Congress to expand the Federal Deposit Insurance Corporation’s authority to impose more stringent penalties on senior executives who mismanage lending banks.

Dispatches

The Books Briefing: Nicole Acheampong writes about the gift of rereading. Up for Debate: Readers weigh in on the freedom and frustration of cars.

Explore all of our newsletters here.

Evening Read

Illustration by The Atlantic

GPT-4 Has the Memory of a Goldfish

By this point, the many defects of AI-based language models have been analyzed to death—their incorrigible dishonesty, their capacity for bias and bigotry, their lack of common sense. GPT-4, the newest and most advanced such model yet, is already being subjected to the same scrutiny, and it still seems to misfire in pretty much all the ways earlier models did. But large language models have another shortcoming that has so far gotten relatively little attention: their shoddy recall. These multibillion-dollar programs, which require several city blocks’ worth of energy to run, may now be able to code websites, plan vacations, and draft company-wide emails in the style of William Faulkner. But they have the memory of a goldfish.

Read the full article.

More From The Atlantic

Israel’s tectonic struggle So where were the regulators? Photos of the week: marsh maze, volcanic ash, dying star

Culture Break

Read. These 10 poetry collections are worth revisiting again and again.

Listen. Our new narrative podcast, Holy Week, is the story of a revolution undone.

Play our daily crossword.

P.S.

Today, the International Criminal Court issued an arrest warrant for Russian President Vladimir Putin and one other Russian official for their possible involvement in the kidnapping of what could be thousands of Ukrainian children. The ICC was created in 1998 by the Rome Statute, an international treaty, and began holding its first sessions in 2003, but it doesn’t have a lot of power: Russia, China, and the United States are not parties to the statute, and neither is Ukraine (which has nonetheless granted the ICC jurisdiction over its territory). A Kremlin spokesperson, of course, immediately waved away the warrant as irrelevant.

Things could get interesting, I suppose, if Putin ever travels to a nation that is part of the ICC, which is almost every other country in the world. Would another state decide to enforce the ICC warrant and arrest a foreign leader? That’s pretty unlikely, but it’s something Putin would at least have to think about if he ever decides to venture too far away from his Kremlin bunker. In the meantime, unfortunately, he and his commanders will continue their crimes in Ukraine, but the ICC warrant is at least a welcome symbolic statement.

— Tom

Isabel Fattal contributed to this newsletter.

The collapses of Silicon Valley Bank and Signature Bank this past weekend were the end point in an all-too-familiar cycle: first the boom, then the breathtakingly speedy bust, and then the bailout. We are now at the postmortem moment—when everyone wonders where the regulators were.

Silicon Valley Bank has already become notorious for how obvious its red flags were. Perhaps the most telling was the rapid growth of its borrowing from the Federal Home Loan Banks system. Banking experts know this Depression-era group of government-sponsored lenders as the second-to-last resort for banks. (The Fed is, as always, the lender of last resort.) At the end of last year, Silicon Valley Bank had $15 billion of FHLB loans, up from zero a year earlier.

“That’s the type of flag that says you need to look closely,” Kathryn Judge, a Columbia law professor who specializes in financial regulation, told me. But there’s no sign the loans triggered any regulatory attention.

Primary responsibility for the debacle lies, of course, with SVB’s management. But regulators are supposed to grasp that they exist because bankers are always tempted to take risks. Bankers want to grow too fast, borrow cheaply, lend freely, and lock their investments up unwisely for long periods in hope of gaining higher returns.

Some commentators are now reiterating calls for banking rules to be tightened, which is probably a wise move. But the collapse of the two banks proves once more that the culture of the regulators is as important as any rules, laws, or tools at their disposal.

[Annie Lowrey: You should be outraged about Silicon Valley Bank]

At least one journalist detected banks’ rising vulnerabilities, including those of Silicon Valley Bank, as early as last November; the Federal Deposit Insurance Corporation’s own chair had also warned about the problem. A few short sellers even started betting against the bank’s stock. Now, however, the combination of reckless bankers and lax regulators has left us with a financial crisis and a federal-government bailout—and the well-rehearsed spectacle of regulators promising to do better next time. (And yes, this was a bailout. Some depositors were facing losses and the federal government, backed by the public, prevented that—at as-yet-unknown scale and cost.)

One troubling aspect of this particular collapse is just how unremarkable a bank run it was, how basic its causes were. Regulators didn’t need any fancy analysis to detect the danger at Silicon Valley Bank. They just needed to notice its financial results. Granted, in 2018 Congress had loosened the post-global-financial-crisis Dodd-Frank regulations that would have required a bank like SVB to undergo more frequent stress tests, but those tests measure exotic or extreme risks. All that was required in this case was regular supervision. The bank had clear risk-control flaws and disclosed losses on its books, right there in its Securities and Exchange Commission filings.

Silicon Valley Bank’s assets had grown dramatically, quadrupling in five years, as had its deposits. Both phenomena are almost always worrying signs. The bank was also overly concentrated in one sector of the economy, and an unusually large proportion of its deposits—about 94 percent—was uninsured, above the $250,000 limit that the FDIC will guarantee per deposit.

No bank can survive if every creditor asks for their money back at once. The larger the portion of a bank’s clients that could wake up one day to realize that their deposits are not protected, the greater the risk of a run.

What Silicon Valley Bank did with those deposits should have been another warning signal. It used them to buy too many long-term bonds. As interest rates go up, bonds lose value. Nobody should have needed the warning, but the bank itself said that interest-rate risk was the biggest hazard it faced. And regulators should have noticed before the bank began borrowing heavily from the FHLB system.

[James Surowiecki: What social media is doing to finance]

In its SEC filings in the third quarter of last year, the bank’s parent company disclosed that it was sitting on losses from its bond purchases big enough to swamp its total equity. That would have been a good time for supervisors to tell the bank to get its act together.

Silicon Valley Bank was far from doing so: It hadn’t had a chief risk officer for most of that year. “Regulators had to know that, and it has to matter,” Jeff Hauser, the founder and director of the Revolving Door Project, a D.C. nonprofit that tracks the regulatory state, told me. “Once we valorize success as proof of wisdom, it’s hard for a lowly bank examiner to say, ‘This place doesn’t have a risk officer and doesn’t have a plan to address the risk on its books.’”

Bank regulators have awesome powers. They can go into a bank, examine its operations, and demand changes. The problem is that they rarely do. “The regulators are like all the conflicted agents in ratings [agencies] and other areas,” Chris Whalen, a longtime financial analyst, told me. “They go with the flow in good times and drop the ball in bad times.”

The San Francisco Fed, which regulated the parent company, and the California regulators, which oversaw the bank itself, could have required SVB to raise capital last year, when it was less vulnerable. They could also have required the bank to increase rates on its savings accounts—in other words, to pay people more to lend it money. That would have eroded earnings but it would’ve kept customers from fleeing. Ask Greg Becker, the bank’s chief executive, today if he would rather have reduced per-share earnings or avoided having superintended the second largest banking collapse in U.S. history.

So why don’t we have regulators who can be relied on to do their jobs?

Part of the answer is a legacy of the Trump administration’s penchant for installing regulators who are opposed to regulation. Donald Trump appointed Randal Quarles as the first-ever vice chair of banking supervision at the Federal Reserve. (The Fed did not respond to questions for this story.) Quarles saw it as his mission to relax the post-financial-crisis regime. He sent unambiguous signals about how he felt about aggressive regulators—“Changing the tenor of supervision will probably actually be the biggest part of what it is that I do,” he declared in 2017. Translation: Any sign of showing teeth and he’ll get out the pliers. And when Jerome Powell was nominated to be the chair of the Fed, in 2017, he told Congress that Quarles was a “close friend,” adding, “I think we are very well aligned on our approach to the issues that he will face as vice chair for supervision.” Naturally, Quarles supported the 2018 law to roll back stress tests—something that Becker himself had called for. Quarles also did not respond to my request for comment.

[Jerusalem Demsas: ‘Financial regulation has a really deep problem’]

This crisis raises the old issue of how strange it is that the Federal Reserve regulates banks at all. In the years leading up to the 2008–09 financial crisis, an alphabet soup of regulators ostensibly shared responsibility for banking oversight along with the Fed: The OTS (Office of Thrift Supervision), the OCC (Office of the Comptroller of the Currency), the SEC (Securities and Exchange Commission), and the CFTC (Commodity Futures Trading Commission). Banks and financial entities played these agencies off against one another to shop for the least restrictive. Policy makers and legislators knew this and toyed with changing the architecture of banking-and-securities regulation. Ultimately, their only action was to close down the least of them, the OTS, and keep the rest, each of which had its own constituency of supporters.

So the Federal Reserve kept its responsibilities. But critics argue that the Fed can never become an effective bank regulator because its chief concern is with the more glamorous business of managing the economy.

The roots of regulatory failure run deeper, however, than the Trump administration’s actions. President Joe Biden’s appointees at the Federal Trade Commission, the Department of Justice, and the Consumer Financial Protection Bureau appear to be trying to wield their powers to make the economy more efficient, safer, and more equitable. But pockets of learned governmental helplessness remain. Regulators have an ingrained fear of stepping in, making people uncomfortable, making demands, and using their clout.

The Fed’s banking supervisors should have been on heightened alert as its governors started boosting interest rates. Silicon Valley Bank faced not only the interest-rate risk to its treasury-bond holdings but also the likelihood of credit losses accumulating on its books from distressed venture-capital firms and declines in commercial real-estate values last year.

The fact that the Fed supervisors weren’t agile with Silicon Valley Bank indicates that they have failed to internalize how woefully fragile our financial system is. The U.S. has suffered repeated bubbles, manias, and crashes since the deregulatory era began under Ronald Reagan: the savings-and-loan crisis, Long-Term Capital Management, the Nasdaq crash, the global financial crisis, the financial convulsions of the early pandemic. Congress and regulators sometimes shore up aspects of the system after the event, but they have failed to foster a resilient financial system that doesn’t inflate serial bubbles. Each time, instead, the regulators reinforce a lesson that if bubble participants huddle as closely together as possible, and fail conventionally, the government will be there to save them.

“One of the most disturbing dynamics here,” Judge, the Columbia Law professor, told me, “is a loss of respect for the Fed as a supervisor, as a regulator.” That is not a good place for the industry’s chief overseer to start rebuilding confidence in the integrity of the American banking system.