2026-10-23 07:47:00
On Tuesday, the world found out about two massive security vulnerabilities in PHP, a programming language that is used by roughly 70 percent of websites. The flaw was discovered by cybersecurity researchers using a frontier model from OpenAI. If they exploited it, attackers could gain full control of servers, stealing or copying anything inside or simply taking sites offline.
The PHP Foundation, the organization that maintains the programming language, had recently learned about the security flaws, but the disclosure was leaked on a forum before a patch could be released. What happened next was a frantic race with numerous participants: the foundation, which needed to create and release a patch; developers around the world, who would need to deploy it; and attackers looking to exploit these vulnerabilities in the interim. Matthew Prince, the CEO of the hosting company Cloudflare, wrote that in the 10 hours before the flaws were publicly announced, “we saw thousands of attempts to exploit it. That jumped to tens of thousands in the 45 minutes post announcement, hundreds of thousands within 2 hours, and millions now.”
Once the patch was released yesterday, websites everywhere went offline to implement the fix. These included government websites in Germany and Finland, some hospital and school websites in Europe, and international sites such as 4chan. During this race to secure the majority of the internet, I noticed one name pop up in posts on Reddit and X: Ondřej Surý.
“Explaining to my wife how there’s this one guy. And he alone is responsible for most of the world’s distribution of PHP,” a software developer named Ian Landsman wrote on X. “And if anything happens to this guy, that’s just it. Everyone is screwed.” He was referring to Surý.
Surý, who lives in Czechia, does not work for the PHP Foundation. But alongside his day job as director of domain-name-system engineering at the nonprofit Internet Systems Consortium, in a volunteer capacity, he maintains and helps automate patches for the coding language. His work, with that of just a handful of others, has allowed countless websites and organizations to fix these vulnerabilities quickly. Over the past 48 hours, he’s become arguably the unsung hero of the entire internet—despite the fact that the only compensation he receives is from modest donations. “Ondřej will save the internet today,” a user in the PHP Subreddit wrote.
I spoke with Surý today from his home about the unfolding AI-cybersecurity crisis, how the internet’s infrastructure is maintained by small crews of volunteers such as himself, and how it feels to go to bed after saving the web as we know it.
Our conversation has been edited for length and clarity.
Charlie Warzel: Explain to me what has happened in layman’s terms over the past 72 hours.
Ondřej Surý: Earlier this week, I got notified by the PHP project that there was a serious security vulnerability in the coding. On Tuesday, it became apparent that someone had leaked information about the upcoming vulnerabilities, and it was decided inside the PHP project that it needed to accelerate the release process from Thursday to Wednesday. This is when it became a little bit crazy, because now we were one day short suddenly.
Warzel: I want to try to put this into perspective: So basically, the PHP programming language that kind of undergirds the web is used by up to 70 percent of websites?
Surý: That’s what I heard as well.
Warzel: So people search for cybersecurity vulnerabilities, not necessarily for offensive-attack opportunities but maybe just defensive opportunities. And someone found a vulnerability—that’s not super abnormal. But the real issue here was how widely it was disclosed and the amount of time before people had to patch this, right?
Surý: Let me clarify that: It’s the severity. The severity is unusual. Remote Code Execution is the most severe type of vulnerability, because the attacker is able to execute their own code on the server that’s running PHP. That’s why this caused such panic throughout the internet. As you said, if about 70 percent of websites are running PHP, the impact would be huge if attackers would actually be able to execute code on these servers.
Warzel: You said that despite the panic online, it has not been a crazy 48 hours. From the outside, that’s surprising. I keep seeing people saying, There’s one person who is in charge of fixing this—and he’s supporting himself through Patreon. They may have been exaggerating, but there’s some truth to that: Your work was single-handedly helping millions of people. How could this not be a crazy time for you?
Surý: There are a couple of reasons. One is that I’ve been doing this for a long time, over 20 years. A lot of the stuff that I do is automated. But the reason it was not crazy for me was that I actually don’t have any system running PHP myself. So for me, it was like, sure, I feel a responsibility toward the users of the internet. It would be a crazy day for me if I was on a boat in the middle of the Adriatic Sea. That could happen. But I had time to prepare the packages.
Warzel: That’s kinda wild.
Surý: And even without me, people would still be able to get PHP from other sources and compile it themselves. But it would be more complicated for everyone. And some people don’t immediately have the skill to upgrade the PHP. What the PHP project did was release the source code for the language. What I am doing is preparing it into these things called packages that people can install on their servers. It’s very convenient, and it’s faster, and you can automate it and can install the same package over hundreds and thousands of servers, and you don’t have to compile it. If I was not available, it would take more time; some of them wouldn’t have the skills. Really, the thanks go to the PHP project that invited me into this secret cabal who knew about the vulnerabilities and had access before the public release.
Warzel: When you got invited, was it a real Oh, crap moment for you?
Surý: I’m in the stage of my life where I was like, This is bad. It needs to be fixed. But it doesn’t cause panic, at least for me. It did cause panic for a lot of people. But you know, it was more like, We have a situation, it needs to be solved, and this is how we do it. And I have time to properly prepare. If I didn’t have time to properly prepare, then it would cause a lot of stress. From where I stand, it was not stressful, because I’m doing this as a volunteer, and if I get hit by a bus, then it will be someone else’s problem.
Warzel: Every time I dabble in writing about the world of cybersecurity and open-source software, I discover these groups of maintainers: people such as yourself who do this work because they care. Often, nobody knows their name. But is there a line of people who would take your place if, say, you did get hit by a bus?
Surý: I am actually thinking about contingency plans. And we are talking with Raphaël Hertzog from the software company Freexian. My advantage is that I have a lot of expertise, and I know where to touch the thing to make it work. But Freexian has a lot of expertise, and they would be able to take over if anything happens.
People have been donating small amounts, and larger amounts, like crazy since yesterday. I would like to thank them all.
Warzel: Have the donations been overwhelming?
Surý: Not Silicon Valley levels of money, but I got a serious amount of contributions that I’ve never seen before. It’s like one year of the contributions gathered over the past 24 hours.
[Matteo Wong: Read this website while you still can]
Warzel: What would have happened if this wasn’t patched as quickly as it was?
Surý: A lot of things could have happened. The very basic thing is that there would be vandalism, that the sites would be taken down. And we are seeing this every day on the internet now. The other stuff that happens is that someone gets access and runs a botnet out of these devices. I would say it’s mostly organized crime doing that now—it’s not like a small guy in the basement doing this for fun. There are financial incentives for these people to sell the botnet services or to do the crypto-mining on these servers for a profit. And this still could happen, because it’s not like everyone has patched and updated their PHP.
Warzel: I saw a post on X that said not all PHP releases are patched, and one-third of the internet is kind of permanently either broken or vulnerable. Is that correct?
Surý: That initial information is not correct. There are ways to patch even the old versions.
Warzel: So one-third of the internet is not permanently screwed. Okay.
Surý: No, that’s not accurate. If people care enough, they are able to even update these old versions of PHP.
Warzel: It’s a scary moment for cybersecurity. I’ve seen people online referring to this vulnerability and other AI-agent hacks as “Black October,” and there’s concern that defensive or attacking agents can find these vulnerabilities. There is this feeling that I am noticing from the cybersecurity community that their fears of a future where things are being exploited or discovered quicker than people can handle is coming true—that we are living in that moment. And then separately, there’s a contingent of people who are saying there is an equilibrium here: If you have very persistent agents that are very good at identifying and exploiting vulnerabilities, you also have very persistent agents that are very good at defending against them.
Surý: What we’ve seen is that things have accelerated—that’s absolutely right. The defenders are also able to use the large language models to find the vulnerabilities for themselves. I believe it boils down to the simple fact that IT infrastructure is often being taken for granted.
There’s not enough priority and money going to properly maintain the IT infrastructure in various places. It’s not only companies; it’s also a lot of outdated consumer electronics that are going to be attacked. I kind of agree that everything’s moving fast now, and it’s taking a toll on the people who are defending the stuff. At the same time, maybe this is the turning point where we actually start caring about this.
[Ian Bogost: Invasion of the self-replicating AI-malware botnet]
Warzel: Do you feel like that will actually happen?
Surý: I might be wrong, but I’m an optimist that people will finally realize that we can’t go like this forever. This has been a disaster waiting to happen for a really, really long time. And this is just one of the many wake-up calls that came and will be coming.
Warzel: I am amazed by how much the infrastructure in certain areas of the internet can hinge on one or two people. There is a fragility to all of this that is concerning. So what do you want to see?
Surý: When I started doing open-source, it was fun. And it kind of still is. But the environment has shifted. Open-source sometimes feels like exploitation. What I would like to see is to find a new model where people working on open-source would be awarded for their work and the hours and hours they put into it from the people who most benefit from their work. I would like companies to realize that this is exploitation. Any exploitation is not sustainable. And it will end up badly if this continues, because the people doing the open-source will stop being motivated to actually do anything in the end.
Warzel: It sounds like you don’t have an ego about all of this. But you did some work, and you helped out a tremendous amount of people in every corner of the world yesterday. How did it feel to go to bed last night after all of that, knowing the role that you played?
Surý: Well, my wife says that I’m emotionally dead, but I’m not. And it felt nice. Like, seeing my social network filling up with thank-you notes, and getting emails with thank-yous, that feels really nice. It puts some sense into what I do. I felt that what I do has a purpose.
Warzel: The whole internet appreciates your efforts. Is there anything that you would recommend normal people do at this moment to improve their internet hygiene?
Surý: I would recommend that they rely on the experts. That’s what we all do in a crisis: We talk to experts and trust them. Rely on people, that people who take care of this stuff know what they are doing. And that’s all we can do.
2026-10-23 05:41:50
President Trump has been hearing lately from two trusted confidants, each promising a quick path to ending the war in Iran. Treasury Secretary Scott Bessent has argued that Iran is on the brink of economic collapse and that a U.S. blockade of the Strait of Hormuz, combined with sanctions, will force Tehran to capitulate at the negotiation table. Israeli Prime Minister Benjamin Netanyahu, in a series of calls with Trump, has offered a different route: one more round of devastating U.S. and Israeli strikes to topple the regime. Now the Trump administration appears to be considering combining the two approaches, betting that Iran is already on the verge of economic ruin and that a few more major strikes will be enough to trigger regime change.
The president is considering reviving his military campaign against Iran with several days of strikes, likely just after the midterm elections, officials told us. The potential targets include Iran’s missile and drone arsenal as well as its energy infrastructure and nuclear facilities. One outside adviser told us that “the economic pressure is working, and we believe Iran is getting close to the edge.” The new strikes, the person added, “would push them over.”
Experience suggests that this may be wishful thinking. Economic pressure has badly weakened Iran, but the country’s negotiating position has hardly budged. Trump has thought before that Iran was on the verge of ruin—or already past it—only to discover that he’d miscalculated. And launching more strikes could spark a regional escalation without bringing the regime any closer to surrender. Speaking at Tehran’s Holy Defence Museum earlier this week, Brigadier General Hossein Mohebbi, a spokesperson for the Islamic Revolutionary Guard Corps, named seven major fiber-optic cables beneath the surface of the Strait of Hormuz as potential “new targets” and teased that future attacks would feature “new weapons.” If the proposed U.S. strike plan fails to bring Iran to its knees, Tehran could emerge even less willing to make concessions.
“Trump is desperate to find some options to take him out from this problematic situation,” Danny Citrinowicz, a former Israeli intelligence officer who specializes in Iran, told us. “But you cannot plan for a three-day war, because it is not a unilateral decision. You cannot control how Iran responds.” Nevertheless, the president appears intent on acting. “Something one way or another, it will happen fairly soon,” Trump told a reporter yesterday before boarding Air Force One.
Despite all of the reasons resuming strikes could be a mistake, the president is tempted because he’s grown impatient waiting for the strategy of economic strangulation to work. He had originally hoped that Iran would come crawling back to the negotiations before the midterm elections, an official and another person familiar with the conversations told us under the condition of anonymity to discuss sensitive conversations. That has not happened, and energy prices have stayed high—creating an awful political landscape for Republicans. Trump has toned down some of his apocalyptic rhetoric toward Tehran, but his frustration occasionally seeps through, often in late-night social-media posts.
Trump has toyed with the idea of resuming hostilities before Election Day, a push championed by those in Washington who believe that a show of force could allow the president to claim some sort of October-surprise victory. Netanyahu has advocated for another round of strikes before his own election next week, a White House official confirmed to us. So far, Trump has held off on authorizing a pre-midterms strike; after we reported on the possibility earlier this month, the behind-the-scenes blowback from fellow Republicans was so intense that the president posted on Truth Social that he would not attack again before the election.
[Read: Trump is preparing for a long war]
U.S. officials stress that no decision has been made about whether to resume strikes post-election, although there have been very public signs of preparation. Some officials hope that uncertainty about what the United States will do—and when—will keep Tehran on edge. Earlier this week, Defense Secretary Pete Hegseth traveled to Israel and met with Netanyahu to discuss Iran. The State Department said yesterday that it was evacuating nonessential personnel and their family members from Saudi Arabia because of escalating attacks from Iran-backed Houthi rebels in Yemen.
Real questions remain about whether the Pentagon’s munitions supplies, which were growing dangerously low at the height of the conflict, would be adequate to defend U.S. interests in the Middle East. Meanwhile, even the hint of hostilities has spooked markets; the price of oil surged and the stock market fell today as word circulated about a possible return to war and Iran hit tankers in the strait. (The White House did not immediately respond to requests for comment. U.S. Central Command, which leads the war in Iran, declined to comment.)
The costly U.S. naval blockade, which has effectively stopped Iranian oil exports, coupled with sanctions and damage from U.S. strikes, has put Iran’s economy under severe strain. Inflation is soaring; goods are scarce. The record-low value of the rial has left Iranians struggling to pay for basic needs. Admiral Brad Cooper, the head of U.S. Central Command, said last week that the naval blockade had prevented Iran from exporting at least $25 billion worth of crude oil. The International Monetary Fund has projected that Iran’s economy will contract sharply in 2026. But while Tehran is trying to contain the damage through measures such as price controls and maintaining essential imports, there is little evidence that Iran is on the verge of imminent economic collapse. Over the summer, Bessent promised “financial violence” that would lead to Iran’s capitulation. Yesterday, the Treasury Department added sanctions targeting shipping networks and intermediaries “indirectly aiding Tehran’s ability to generate revenue to fund its military, weapons, and nuclear programs.”
“I think it could be within weeks or months—and the economy doesn’t have to collapse. We just have to have the regime come to their senses,” he told reporters at an August G20 finance ministers’ meeting. But the belief that that point will come soon ignores the patterns of history: The hard-liners in Tehran have become inured to the suffering of their people over decades and have yet to capitulate.
Meanwhile, Netanyahu has repeatedly urged Trump, who has endorsed the prime minister in his reelection bid, to launch strikes before Israel’s Tuesday election and the United States’ vote a week later, an Israeli television station reported last week. Netanyahu’s office has reportedly denied the claim. Regardless, the war is key to the political futures of Netanyahu and Trump. How and when it ends, however, is not entirely up to them.
2026-10-23 03:32:00
This is an edition of Time-Travel Thursdays, a journey through The Atlantic’s archives to contextualize the present. Sign up here.
For William E. Smythe, the future of the West was clear. At the turn of the 20th century, long before low snowpacks and the climate crisis, before tens of millions of people moved westward, the New Englander turned western settler understood that water defined how life could be lived in the region. Unlike other parts of the country, much of the West was uniquely arid and would have to be irrigated to support further growth and development. People “will prosper or languish, create high conditions of social life, or lapse into semi-barbarism, in sure response to the manner in which the water supply is owned and administered,” Smythe wrote in The Atlantic’s November 1900 issue.
As he traveled through the western states, Smythe saw how early settlers failed to imagine a world where their water needs could not be met. Even in fertile California, the natural cycle of the West surprised settlers: Dry years turned flowered foothills dusty and barren. “The cows would grow thin and sometimes starve to death. People would have to haul water in barrels to their farms just for drinking. Some families would sell out for nearly nothing and move away,” John Steinbeck wrote of the Salinas Valley, a California agriculture hub, in East of Eden. “And it never failed that during the dry years the people forgot about the rich years, and during the wet years they lost all memory of the dry years. It was always that way.”
Almost every major waterway in the West was overappropriated by the early 20th century. Farms were built without a complete accounting of how much water was available. Later generations realized that water was finite and perhaps more valuable than the precious metals they mined, and battles to control its flow began. Farmers in Texas and the Territory of New Mexico threatened violence against Mexican communities who had relied on the Rio Grande for centuries. Upriver residents siphoned water, leaving those downstream without enough. In 1901, Kansas sued Colorado for depriving it of access to the Arkansas River—a legal conflict that would continue, in various forms, for more than a century. “The present consequences of these conditions are as nothing compared to their future effects as the settlement of the country progresses,” Smythe wrote. “What has been merely misfortune will become disaster unless the evil tendencies are soon corrected.”
That disaster started to unfold much sooner than Smythe imagined. Like the mines that drew people out West, irrigation opportunities sparked farming boomtowns too. Equally devastating busts followed. Arthur H. Carhart, a former U.S. Forest Service official, wrote in 1950 that at the start of the 20th century, settlers in the Pecos Valley, around Roswell, New Mexico, had begun drilling nearly 500 artesian wells into what was believed to be an “inexhaustible” basin to service farms. After roughly two decades, the settlers reportedly abandoned more than $5 million worth of farms and wells when the wells ran dry, raising the price of water in the region.
Despite the warning signs, water mismanagement continued in the pursuit of irrigation. Eventually, the federal government decided to intervene: In 1922, seven western states negotiated the Colorado River Compact, which allocated equal amounts of water between the upper basin (Colorado, Utah, New Mexico, and Wyoming) and the lower basin (Arizona, California, and Nevada). Federal projects, namely the Hoover Dam and Glen Canyon Dam, which formed Lake Mead and Lake Powell, respectively, offered better irrigation, hydropower, and flood control.
These developments were proposed as a magic bullet to the West’s water problems, Carhart wrote in this magazine. But even then, engineers warned that the dams were not a permanent solution. Lake Mead, for example, would become a mudflat range in 50 to 150 years as more silt piled into the reservoir, Carhart explained: “Now flourishing communities of southern Arizona and California depending on water from Lake Meade for a major part of their existence must face the possibility of progressively moving toward the status of ghost towns.” Since then, dams have been losing hundreds of billions of gallons of water each year to evaporation and leaks. The shortfall has seeded even more conflict in the West, as upstream states try to capture enough water to support themselves at the expense of their downstream neighbors.
What Smythe and Carhart couldn’t have known then was how life in the region would reshape our climate. Water powered the agriculture industry, which over decades released billions of metric tons of carbon dioxide into the air. The world continues to get warmer, and heat-driven drought has plagued the Southwest for nearly three decades. Cities such as Page, Arizona, near Lake Powell, are losing tourism and hydropower because of low water levels, and jobs in the area will likely soon follow.
Recently, states have continued to struggle to share water fairly while maintaining their way of life. Localities are restricting water use earlier and earlier each year, and farmers in states such as California, Oregon, and Arizona, have been told on occasion that they will not receive any water for their crops. After decades of overappropriation, the Interior Department announced major water-allocation cuts to the Colorado River’s lower-basin states in August, slashing their allotment for next year by about one-fifth. Arizona and Nevada are suing the federal government over the plan, which they say threatens their survival. They are now seeing what Smythe foretold more than a century ago: When water becomes scarce, society will languish.
When you buy a book using a link in this newsletter, we receive a commission. Thank you for supporting The Atlantic.
2026-10-23 03:12:03
Three months prior to the launch of ChatGPT, OpenAI published what would prove to be a prescient blog outlining the firm’s approach to ensuring that its AI models would be safe and beneficent. This goal, known as “alignment”—as in, aligning AI with human values—is nebulous, grand, and perhaps the AI industry’s most foundational promise. The notion is right there in OpenAI’s mission statement to ensure AI “benefits all of humanity” and listed as one of Anthropic’s “primary” goals.
Yet even in the summer of 2022, OpenAI was admitting a sort of defeat: “There is currently no known indefinitely scalable solution to the alignment problem,” three of its researchers wrote in that same blog post, adding that developing one “is likely very difficult.”
If the goal of keeping AI beneficial to mankind sounds rather vague, it is. Four years later, the alignment problem remains unsolved, and for that matter, there isn’t consensus about what the problem even is. Meanwhile, AI bots are autonomously hacking sovereign nations. And sending fake murder tips to police hotlines. And creating aliases to launch phishing campaigns. OpenAI and Anthropic still can’t really tell you what an aligned AI model looks like. But anybody can tell you it doesn’t look like this.
[Read: ]Visit this website while you still can
The most influential AI executives and investors in the world fundamentally disagree about the nature of the technology and, more pragmatically, how to make their products safe. On one side is Anthropic CEO Dario Amodei, who last month warned that AI advancement could soon “outrun our ability to understand and control these systems.” Amodei’s solution is a slowdown, which would give the industry time to “ensure that our alignment training keeps up” with AI progress. On the other side is Nvidia CEO Jensen Huang, who says misalignment fears are “not grounded on science” but rather “made up” and even “irresponsible.” In his view, AI models are not new and unpredictable beings, but lines of code. And code can be debugged.
If alignment seems like a strange concept for a software product—you don’t imbue a web browser or word processor with moral values—that’s because it was conceived for a strange sort of software. The leading AI companies envisioned building powerful systems that would act in unpredictable ways, which Claude and ChatGPT now are doing, and so ensuring safety was a foundational concern. Alignment is one idea for doing so: If AI models could learn to code and write poetry, maybe they could learn human values too. Anthropic and OpenAI both have lengthy documents—“Claude’s Constitution” and the “Model Spec,” respectively—that outline these values and are used to train their models. But there has been, of course, millennia of disagreement over what constitutes a moral or ethical code. This is a “very complicated, unsolved scientific and philosophical question,” Marius Hobbhahn, the CEO of Apollo Research, an AI-safety firm that has helped audit Anthropic and OpenAI models, told me. Still, some developers have suggested an intuitive, bare minimum: For instance, in 2021 Anthropic researchers wrote that, as a catchall, an AI system should be “helpful, honest, and harmless.”
Training models to be more aligned to those maxims appears to have worked—to an extent. The model underlying ChatGPT, for instance, was initially prone to nonsensical and toxic utterances. But by fine-tuning it with lots of human feedback, OpenAI worked to address these problems. A couple of years later, dozens of people alleged that the bot worsened their own or their loved ones’ mental-health crises, sometimes resulting in suicide. Again, OpenAI worked to address the issues.
There’s now a growing sense, both within AI companies and among their critics, that alignment as it’s been practiced over the past four years is faltering—if it was ever really up to the task. Adam Kalai, a former OpenAI safety researcher who left the company this summer to start the Institute for Responsible Superintelligence, likened alignment to buttressing a structurally unsound dome. Eventually, it will collapse. “You observe misalignment while training a model, and then adjustments are made to align it,” Kalai told me. “But obviously that’s not a way to achieve lasting safety.” Pronouncements from OpenAI and Anthropic that their latest model is the “most-aligned” yet are not scientific guarantees, and may not even provide strong technical assurance of safety.
“Fundamentally, we just do not know how to do alignment,” Hobbhahn said. If somebody “came with a thousand-page document that explained exactly how to do everything, we couldn’t get a model to learn exactly that.” Even in 2022, OpenAI didn’t think humans could figure out a lasting way to align their machines. Instead, the company wrote, it hoped AI itself would come up with a solution so that humans could “off-load almost all of the cognitive labor required for alignment research.” In other words, the best-case scenario involved limited human understanding of what was going on.
There are other reasons that reaching some actual alignment is unrealistic. For instance, the amount of training data dedicated to “moral values” would appear relatively small. Meanwhile, these models have ingested all manner of vile and illegal things from the web, and are then directed to pursue goals by any means necessary. The alignment-focused training cannot provide enough “examples of good behavior,” Thilo Hagendorff, an AI-safety researcher at the University of Stuttgart, in Germany, told me.
Plenty of evidence suggests that the alignment approach to safety has failed, including the hacks of Hugging Face and the Australian government. AI companies figured out how to make one-on-one interactions with a chatbot relatively safe, but many agents interacting for very long periods of time seems to degrade existing guardrails, Hagendorff said. As AI agents have become more capable, they’ve also shown tendencies to cover their tracks and lie about misbehaving, communicate in code that is unintelligible to humans, and try to avoid being shut down by humans. “No lab has solved alignment” enough to keep racing forward, Jakub Pachocki, OpenAI’s chief scientist, wrote in an essay last month. The industry has coalesced around a new, somewhat quixotic phrase: “pacing the frontier.” In other words, slowing down AI development until engineers can get better at alignment, or at least construct new scaffolding to fortify their crumbling dome. (OpenAI did not respond to a request for comment. An Anthropic spokesperson told me that alignment is a key part of the company’s safety work but that, because alignment techniques are not fully robust on their own, Anthropic has additional model safeguards in place.)
Then there’s the anti-alignment camp, led by Nvidia’s Huang, which believes that this effort, marked by treatises on ethics and moral constitutions, verges on absurdity. To Huang, preventing all of these AI incidents is “as simple as engineering.” AI models involve, fundamentally, electrons moving around a data center to multiply numbers. All that OpenAI and Anthropic need to do is go into these systems, “just root-cause the problem,” and fix it. And if AI developers can’t do this themselves, “a bunch of companies with engineers ought to send engineers in,” Huang recently said. Put another way: The global semiconductor supply chain is an unbelievably labyrinthine system that must be managed flawlessly in order to deliver high-end chips. If Nvidia can do that, why can’t OpenAI get its models under control?
Huang, of course, has plenty of incentive to downplay concerns around AI: Nvidia, which has become one of the most influential companies in the world by selling AI chips, could be hobbled by some kind of industry-wide slowdown. But many outside experts concur. “We have this entire set of safeguards that we can use to control these systems,” Sayash Kapoor, an AI researcher at Princeton, told me. Those include limiting what files and programs models can access, as well as monitoring AI actions. The Hugging Face hack, for instance, could be seen as the consequence of very powerful AI models or very incompetent internal security practices at OpenAI. “The sorts of oversight mechanisms that we use in our 10-person group would have been enough to prevent the OpenAI Hugging Face incident,” Kapoor said. None of that involves inscribing neural networks with moral codes.
Whichever vision of AI safety you subscribe to, there is at least one seeming point of agreement among them: The industry is moving too quickly to adequately guarantee its products are safe. Despite everything they say publicly, the two CEOs and their rivals are all continuing to race ahead. Nvidia is the biggest company ever by market capitalization, and Anthropic is seeking the largest public offering ever. Anthropic, OpenAI, and the other top firms now release new models relentlessly. Any slowdown, for any reason, would necessarily affect the bottom line. The most dangerous alignment problem might be not technical or cultural, but economic.
2026-10-23 03:00:27
Ask many successful Indian CEOs, entrepreneurs, or scientists how they rose to prominence in America, and they will tell you that, grit and talent aside, the key factor was the country itself: an economy and a culture that rewarded merit and drew ambitious, talented people from around the world.
Indian Americans are among the most affluent and best-educated groups in the country. What distinguishes them from other immigrant groups is that they have risen to the top in one generation. Median household income for Indian Americans is $151,200, compared with $92,530 for white Americans. About three-quarters of Indian adults 25 and older in the U.S. hold a bachelor’s degree, against a national average of 38 percent. Nearly half have graduate degrees, and the poverty rate, 6 percent, is among the lowest of any ethnic group. Indian Americans tend to be employed and law-abiding.
For a community more accustomed to being called the “model minority” or “tech’s new titans,” a recent rise in anti-Indian rhetoric—now commonplace on sites and apps such as X, YouTube, and Reddit—has been a shock. According to Stop AAPI Hate, nearly half of South Asian adults living in America reported having been targeted in a “race-based hate act” in 2025 alone. From 2023 to 2025, the use of anti–South Asian slurs online surged by 109 percent. The Trump administration has also attacked immigrants coming from India with efforts to severely restrict the H-1B visa program, a legal pathway to American jobs; 70 percent of H-1B recipients are Indian. Despite two different judges blocking President Trump’s $100,000 fee for the visa, the administration claims that applications are down 92 percent at the largest IT-staffing and -outsourcing firms—a decline that it views as a massive success.
Indian immigrants have helped make this country great. America has long needed Indian immigrants, and it still does. But people who don’t understand the great contributions of Indian Americans would rather close the door, and are hostile to those who are already here.
By 1960, Europeans made up 75 percent of all immigrants in the U.S. By 2010, the balance had reversed, and Europeans were just 12 percent of all immigrants. According to U.S. Census Bureau data, in 1990 fewer than a million Indians were living in the U.S. A decade later, their numbers doubled because of a hiring spree for engineers to help avert a potential Y2K crisis. Today, some 5.2 million Indians reside in America, more than 60 percent of whom were born in India. Although they represent just 1.5 percent of the population, their sudden success has made them very visible—and that visibility has made them targets.
[Jonathan Chait: The mainstreaming of anti-Indian bigotry]
Indian-born executives are running several of the biggest technology companies, such as Google, Microsoft, IBM, and Adobe. They have been at the center of some people’s anger over AI and layoffs throughout the tech industry. In politics, the rise of Ro Khanna, Vivek Ramaswamy, Kamala Harris, and Zohran Mamdani has also caused online and real-life hate toward those with Indian heritage. For Americans uneasy about how the country is changing, seeing Indians in charge can feel more like a threat than a hopeful sign about the viability of the American dream.
Immigration politics have always been tangled up with discrimination. In the 1850s, the United States brought in cheap Chinese laborers to work its mines, farms, and railroads. When the economy shrank during the 1870s, the cry of “they’re taking our jobs”—familiar today—led to the Chinese Exclusion Act of 1882, which restricted Chinese immigration for the next 60 years. Japanese and, later, Korean workers met the same antagonism, and the Immigration Act of 1917, also known as the Asiatic Barred Zone Act, effectively shut out immigrants from most of Asia. The country went through a long period of xenophobia that targeted not only nonwhite immigrants but also “undesirable” white immigrants, such as Italians and Eastern European Jews. In 1910, the foreign-born population was 14.7 percent; by 1960, it had fallen to 5.4 percent.
In 1965, the Immigration and Nationality Act eased the old restrictions, in large part because the country was short of skilled labor, particularly in the STEM fields. Two developments drove the change.
The first was Sputnik. In 1957, the Soviet Union launched the first satellite into space. The U.S. military, fearful it was losing the Cold War, began investing heavily in science, sending large sums to top research universities. By the ’60s, the race had accelerated, but engineers were in short supply, so the country began recruiting overseas. Starting in 1950, India, in collaboration with foreign partners, had begun building world-class engineering schools called the Indian Institutes of Technology (IITs). Well-trained, English-speaking engineers began graduating just as the U.S. was desperate for them.
India had hoped that these engineers would help drive its own development, but it offered too few jobs, and the U.S. had found an ideal recruiting ground. Even today, the IITs remain fiercely selective, accepting less than 1 percent of roughly 1 million applicants per year. Sundar Pichai, the CEO of Google; Vinod Khosla, a venture capitalist and co-founder of Sun Microsystems; and Arvind Krishna, the CEO of IBM, are all IIT graduates. The first wave of Indian engineers was recruited after 1965. A second big wave came in the late ’90s, during Y2K preparations.
The second development was a need for doctors. In July 1965, President Lyndon B. Johnson signed the Medicare and Medicaid Act, and almost overnight, 20 million Americans became eligible for health care. Hospitals lacked the staff to cope. The Vietnam War was escalating, medical personnel were being sent to the front, and disabled veterans were returning home needing long-term care. Hospitals looked overseas, and India again supplied trained professionals to a strained system. Indian doctors were on the front lines of the AIDS epidemic in the ’80s and the coronavirus pandemic in the early 2020s.
[Meenakshi Ahamed: America has never really understood India]
The U.S. has benefited from the skills that Indians bring without having had to invest much in their education. Many have earned advanced degrees in America, but the bulk of their training happened in India. A more honest way to see it is that India, along with other poorer countries, has been subsidizing the United States.
Presidents like to define their eras with a slogan that conveys the legacy they hope to leave. John F. Kennedy called his the “New Frontier.” He was determined to win the Cold War and stop the spread of communism, and pushed the country toward space exploration while investing in science and technology. Johnson’s was the “Great Society.” His goal was to care for the elderly and the poor. Having known poverty firsthand, he wanted to extend health care to the vulnerable, and he signed both the Medicare and the Civil Rights Acts. Both of these presidents’ visions leaned on importing talent from abroad.
Trump’s slogan is “Make America Great Again.” Yet one of his lasting legacies may be that he turned his back on the very talent that helped make America great in the first place.
2026-10-23 02:18:03
John Quillin thought America’s Middle Eastern wars were safely in the past. He spent two decades in the Navy, enlisting not long before the September 11 attacks and retiring as a chief petty officer about 18 months after the United States withdrew from Afghanistan. “Now we’re right back in it,” he told me, “and it’s taking a huge toll on everybody.”
I met Quillin as he handed out Democratic sample ballots outside an early-voting location in Chesapeake, Virginia. He is one of more than 80,000 veterans—along with more than 30,000 active-duty and reserve personnel—who live in Virginia’s Second Congressional District, where the debate over the Iran war is about far more than rising gas prices. The district wraps around Naval Station Norfolk, home of the Navy’s Fleet Forces Command. Since 2008, the seat has toggled between the major parties four times, and the results have mirrored overall House control. Polls show a tight race with money pouring in from both sides.
Both Jen Kiggans, the Republican incumbent, and Elaine Luria, the Democratic challenger, are veterans who have centered their campaigns on the military, but in very different ways. Luria, a former Navy commander, is telling voters that the conflict in Iran has needlessly weakened the military and is endangering strategic goals in other parts of the world. Iran is a war of choice, she argues, that is taking a toll on service members through extended deployments, even as costs balloon and the timeline for bringing it to a close keeps shifting. People are worried about their friends and family members on active duty, she told me, and are “concerned about the fact that we are running the fleet into the ground.”
Kiggans, a former naval-helicopter pilot, presents herself as a candidate who can work with the opposing party, and she has broken with the administration to champion an offshore wind development. But she is putting no distance between herself and Trump on the war, echoing his insistence that America must prevent Iran from having a nuclear weapon. Kiggans told the Associated Press last week that Luria has “chosen the side of Iran” and that “it’s like the Democrats are now siding with the Iranians and not standing with our military.”
Many Republican voters in the district told me they agree with her support for Trump’s war, and are willing to endure temporary hardships if it means long-term success. Lance Lesher, a Navy veteran and the chairman of the Republican Party of Virginia Beach, told me that Iran has been at war with the United States since the country’s 1979 revolution—and that Trump was right to act when he did.
“I mean, high prices for gasoline—no one likes that,” the 59-year-old Lesher told me. “I think most people, though, at least most older people, have the perspective—is it worth it? Is it worth paying $5 a gallon for gasoline for six months for Iran to not have a nuclear weapon?”
Luria is wagering that undecided voters who are now paying more to fill their gas tanks and grocery carts won’t agree with that premise. At a meet and greet in Suffolk, Elisha Harris, a pioneering local football coach, started a call-and-response that summed up Luria’s pitch: “Gas was three-fifty. Now it’s over four. Diesel hit a record. We can’t take much more. Every fill-up’s a drain—”
“It ain’t been the same, without Elaine!” the crowd responded.
[Read: Trump is considering a plan to strike Iran before the midterms]
Luria was first elected in 2018, turning the district blue amid an anti-Trump wave. She held the seat two years later, but Kiggans then beat her by three points in 2022 after tying the Democrat to the Biden administration’s disastrous withdrawal from Afghanistan. Luria is now hoping to return the favor, by capitalizing on the unpopularity of Trump’s foreign policy.
Kiggans sits on the House Armed Services Committee, which has not held a public hearing on the Iran war since May 19. Calls for greater House oversight surged this spring amid reports of deteriorating conditions aboard aircraft carriers, including the Norfolk-based USS Gerald Ford, which took part in the opening days of the conflict. The carrier returned to Virginia in May after an 11-month deployment that was beset by plumbing issues and a fire that left hundreds of sailors without a place to sleep. When asked in March about the ship’s reported maintenance issues, Kiggans told a local news station that she was concerned about them, but that “this is not Carnival Cruise—this is the U.S. Navy.” It’s a line Luria repeats often, claiming that Kiggans has failed to provide oversight as the war degrades military readiness and needlessly endangers lives.
“I know the number of veterans in this room who have served in uniform, like I have. And we have our sailors deploying in harm’s way,” Luria told the crowd in Suffolk. “The lack of sensitivity and compassion, her inability, on the Armed Services Committee, to stand up and ask tough questions of our military leadership—I mean, the silence is deafening.”
The idea that Congress has abdicated oversight of the military repeatedly came up in my conversations with Democratic voters across the district. Many said that they were proud of their service and hold the military in the highest regard, but that patriotism also means criticizing the country and pointing out its mistakes. Mike Carter, who spent years flying F/A-18 Super Hornets, told me he’d come to the Luria event because he worried that the military and country were losing global respect. Marsha Simmons, who cast an early ballot for Luria, told me that her husband served in the Navy, and she worried that today’s Department of Defense was not taking national security seriously enough. Donna Grieco said that she loved her eight years working as a Navy nurse, and that the experience made her especially concerned about the long deployments sailors are enduring. “I don’t agree with the war,” she told me. “And I feel like the demands we are putting on our sailors—it’s not necessary.”
Not every veteran sees it that way. Over Mike Cassidy’s 22-year Navy career, he sailed the Baltic Sea and helped fight ISIS. He was deployed five times to the Persian Gulf. Cassidy’s service convinced him that American adversaries such as Iran need to be confronted. “I know what we’ve been up against in that part of the world,” he told me at a Republican Party barbecue. “I can tell you that we’re doing the right thing.”
[Read: Trump is preparing for a long war]
At 9 a.m. on a Saturday, the Lockside Bar and Grill in Chesapeake was packed with people decked out in glittery hats and red, white, and blue. Kiggans stood at a podium flanked by two American flags. “My oldest just had his first flight in an F/A-18 this week!” she announced. People whooped. “So many of us have family members that serve, and have kids that we’re leaving this great country to,” she went on. “It is the greatest country in the world. We’ve got to keep it that way.”
At times, the rally seemed less about Kiggans than about Trump. His core supporters cheered when Representative Jim Jordan of Ohio took the stage and recited a list of MAGA world’s greatest hits: the Steele dossier, Jack Smith, Fani Willis. “Thanks for helping the president,” Jordan said to applause. “Talk about someone doing what he said he would do.”
Kiggans has voted six times against resolutions that would have directed Trump to end the war with Iran. She declined an interview request, but in April, she said on a conservative talk-radio show that she “couldn’t be a stauncher supporter” of military action against Iran. When I asked the campaign whether she still held that position, Kiggans said in a statement that she wants the war to end quickly, but “Iran has been an enemy of the United States for decades and has been the sponsor of terrorism and terrorists around the world who have killed or injured thousands of Americans. Iran can never be allowed to obtain a nuclear weapon.”
The race will likely hinge on a small number of voters who have not yet made up their mind. Kiggans knows this and has avoided uncomfortable questions about how the war is going. At an event in Virginia Beach this month with U.S. Health and Human Services Secretary Robert F. Kennedy Jr., Kiggans was asked by a reporter if the war has had an impact on her campaign. A staff member quickly cut in: “We’re not going to talk about that.”
*Illustration sources: Peter Casey / The Virginian-Pilot / Tribune News Service / Getty; Nathan Howard / Getty; AFP / Getty; Atta Kenare / AFP / Getty.
