2025-11-29 21:30:00
全球凶杀率正在下降，但究竟是为什么呢？| SimpleImages/Getty Images
一个好消息的来源——我本人和显然也受到风险投资家青睐——是所谓的“叙事违反”（narrative violation）。所谓“叙事违反”，指的是人们普遍认为某件事是这样，但实际证据却显示并非如此。而很少有叙事比“暴力犯罪总是上升”这一信念更常被违反。2023年，IPSO对30个国家的人进行了一项调查，结果显示70%的受访者认为世界变得更加暴力和危险。在美国，自上世纪90年代初以来，多数人几乎每年都告诉民调机构，暴力犯罪在上升。其他调查也表明，许多人坚持认为50年前的生活比现在更安全、更好。
然而，事实却与这种叙事相反：当你实际查看凶杀数据时，会发现世界整体上正在变得更加安全，无论是与遥远的过去相比，还是与本世纪初相比。今年早些时候，我曾写过，上世纪90年代实际上是美国暴力程度极高的十年，而今年美国的暴力犯罪率可能接近历史最低水平，尽管许多美国人——包括总统——仍坚持认为情况并非如此。
最近，世界银行更新了数据，从全球视角来看，这一趋势更加明显。2000年至2023年间，全球凶杀率从每10万人约6.9起降至2023年的约5.2起，这意味着任何随机个体被谋杀的可能性下降了四分之一。尽管全球人口在这段时间内有所增长，但凶杀总数仍有所上升。如果凶杀率没有下降，而是保持不变，那么在这段时间内，将有约150万人额外丧生。这相当于整个费城的人口都因世界变得更加和平而得以存活。
过去并非总是如此。由于好莱坞的影响，我们对古代的暴力有某种想象，但实际情况可能更糟糕。研究者如史蒂芬·平克（Steven Pinker）的工作帮助我们拼凑出中世纪和早期现代时期的暴力图景——在很多地方，暴力程度非常高。最近，犯罪学家曼努埃尔·艾森纳（Manuel Eisner）利用验尸官记录，绘制了14世纪英国伦敦、约克和牛津镇的每起已知杀人事件。艾森纳发现，伦敦和约克的凶杀率在每10万人20至25起之间，而牛津——欧洲最古老的大学所在地——的凶杀率则高达每10万人100起（原因可能是中世纪的牛津学生喜欢喝醉并互相决斗致死）。如今，牛津的大学生最多可能只用到一把小刀，而2023年结束前，该市仅有两起凶杀案。至于伦敦，2025年前九个月的凶杀率低于每10万人1起，这是自2003年开始记录以来最低的水平。这些世纪的变化，可以用一个词来概括：文明。更强大的国家垄断了武力，法院取代了血亲复仇，宗教和哲学运动使残忍变得不正常，而城市商业的兴起使得稳定的合作比无序的掠夺更有价值。暴力不再被视为日常可接受的工具，社会规范也逐渐跟上。
当然，批评者对平克及其同事的乐观观点提出了质疑，尤其是关于战争和殖民暴力方面。但不可否认的是，西方的普通凶杀案比以往少得多。
这种下降并非仅限于西方。多年来，巴西每年记录的凶杀案超过5万起，全国凶杀率在每10万人20多起。然而，巴西公共安全论坛最新报告指出，2024年的凶杀案降至约4.4万起，这是自2012年以来的最低水平，比之前的高峰下降了约25%。报告作者认为，这一下降得益于多种因素，包括联邦政府对安全的重新投入、对平民持枪的更严格限制、帮派之间的停战协议，甚至还有人口老龄化。
但这并不意味着工作已经完成。当今的暴力负担高度集中。2021年，美洲和非洲的凶杀率分别为每百万人口约150起和127起，远高于欧洲或东亚。在这些地区，只有少数国家和城市承担了大部分的凶杀案。例如，海地的太子港或墨西哥的科利马，这些城市某些区域的凶杀率已达到每10万人数百起。全球平均水平的改善并不意味着所有地区都变得安全，某些社区仍然危险。
全球凶杀率下降的研究并不完美，没有单一的“万能钥匙”，但一些模式反复出现。国家基本能力的提升有助于减少凶杀，例如运作良好的法院、较少腐败的警察和可预测的法律体系，使得谋杀更难得逞。有针对性、数据驱动的警务，专注于小范围的热点区域和少数导致严重暴力的人群，似乎比无差别镇压更有效。武器政策、经济和社会状况也起着重要作用。例如，美国的研究发现，家庭和经济困境的指标与凶杀、自杀和毒品死亡率之间存在强相关。当这些压力减轻时，暴力也随之减少。
最后，还有一个因素是所有人都无法控制的，但可能是最重要的：老龄化。年龄是暴力犯罪最有力的预测因素之一，凶杀案几乎全部由（针对）年轻男性所实施。一项2019年的研究发现，自1960年代以来，世界大部分地区15至29岁人口的比例都在下降，而这种老龄化现象解释了近期凶杀率下降的很大一部分。当社会老龄化时，犯罪率往往会下降。全球人口结构的变化——更少的儿童，更长的寿命——似乎正在悄然使人类社会更加和平。
如果你像我一样喜欢通过“叙事违反”来看世界，那么这个趋势值得牢牢把握。有些人认为我们正滑向混乱，但数据表明，尽管缓慢且不均衡，我们正在让彼此互相伤害变得更加困难。发现“叙事违反”很有趣，但构建准确的叙事则更加重要。本文最初发表于《好消息》（Good News）通讯，欢迎订阅！
One source of good news — favored both by me and, apparently, venture capitalists — is what’s known as a “narrative violation.” A narrative violation occurs when everyone thinks one thing, but the actual evidence suggests the opposite.
And few narratives are more persistently violated than one common belief: “Violent crime is always going up.”
A 2023 survey from IPSOS of people in 30 countries found that 70 percent of respondents thought the world was becoming more violent and dangerous. Here in the US, majorities have told pollsters almost every year since the early 1990s that violent crime is going up. And other surveys indicate that many people around the world insist that life was better and often safer 50 years ago than it is today.
So, that’s the narrative. Here’s the violation: When you actually look at data on murder, it shows that the world has largely been getting safer, both as compared to the more distant past and in this century. I wrote earlier this year about how the 1990s were actually an extraordinarily violent decade in the US and how violent crime in the US this year may be headed towards record lows, even as many Americans — including the President — insist it isn’t.
Now, recently updated data from the World Bank looks at the picture from a global perspective and finds something astonishing. Between 2000 and 2023, the international homicide rate fell from roughly 6.9 deaths per 100,000 people to around 5.2 per 100,000 people in 2023. That translates into around a one-quarter decline in the chances that any random person will be murdered.
Because the global population has increased since 2000, the total number of murders has gone up over these years. But, had the global homicide rate not experienced this decline and instead stayed steady, some 1.5 million additional people would have been murdered over these years. That’s equivalent to the population of Philadelphia still breathing because the world has gotten less violent.
We all have a vision of violent antiquity thanks to Hollywood, but how bad was it really? Thanks to the work of researchers like Steven Pinker, we’ve managed to piece together a picture of violence in the medieval and early modern eras — and wow, in a lot of places, it was very high.
A recent project by the criminologist Manuel Eisner used coroner records to map every known killing in the 14th-century English towns of London, York, and Oxford. Eisner found that the homicide rates in London and York clocked in at between 20 and 25 per 100,000 people, while in Oxford, home to the most venerable university in Europe, it was around 100 per 100,000 people. (Why? Apparently medieval Oxford students really liked to get drunk and fight each other to the death.)
Today, the most lethal thing an Oxford undergraduate might wield is a cutting remark; there were all of two homicides total in the city for the year ending in September 2023. For its part, London’s homicide rate was less than 1 per 100,000 through the first nine months of 2025 — the fewest murders since monthly records began in 2003.
What changed over those centuries is, in a word, civilization. More powerful states maintained a monopoly on force, courts replaced blood feuds, religious and philosophical movements de-normalized cruelty, and the rise of urban commerce made stable cooperation more valuable than lawless predation. Violence stopped being an acceptable everyday tool, and norms slowly caught up. Critics have quarreled with Pinker and his colleagues on just how far that optimism should stretch, especially for war and colonial violence, but it’s indisputable that ordinary homicide in the West is far less common than it once was.
It’s not just the West. For years, Brazil recorded more than 50,000 killings annually, with national murder rates in the high 20s per 100,000 people. Yet, a new report from the Brazilian Forum on Public Security finds that homicides fell to about 44,000 in 2024, the lowest level since 2012 and down roughly 25 percent from that earlier peak. The authors credit a mix of factors, including a renewed federal security push, tighter rules on civilian gun ownership, truces between rival gangs, and even demographic aging.
None of this means the work is finished. The burden of violence today is highly concentrated. In 2021, the Americas and Africa had homicide rates of roughly 150 and 127 per million people, respectively — many times higher than Europe or East Asia. Within those regions, a relatively small group of countries and cities bear an outsize share of the killings. Think of Port-au-Prince in Haiti or Colima in Mexico, where recent homicide rates in some parts of these cities have reached well into the triple digits per 100,000 people. The global average can improve even while particular neighborhoods remain terrifyingly dangerous.
The research behind the global murder decline is messy, and there is no single magic lever, but several patterns recur. Improvements in basic state capacity help; functioning courts, less corrupt police, and a predictable legal system make it harder to get away with murder. Targeted, data-driven policing that focuses on small hotspots and the tiny fraction of people responsible for most serious violence appears more effective than indiscriminate crackdowns. Policy choices around weapons matter, as do economic and social conditions. Studies of US counties, for instance, find strong links between measures of household and economic distress and death rates from homicide, suicide, and drugs. When those stresses ease, violence tends to do the same.
One last factor is out of everyone’s control, but it might be the most important: aging. The single most robust predictor of violent offending is age, and homicide is overwhelmingly committed by (and against) young men. One 2019 study found that, since the 1960s, most regions of the world have seen a decline in the share of their population aged 15-29, and that this aging accounts for a significant share of the recent decline in the homicide rate. When societies age, crime falls, all else equal. The global demographic transition — fewer kids, longer lives — seems to be quietly pacifying humanity.
If you, like me, enjoy seeing the world through narrative violations, this is a big one to hold on to. Some people would have you believe we are sliding toward chaos. The numbers say that, very slowly and unevenly, we have been making it harder to kill each other. Spotting a narrative violation is fun; building accurate narratives is even better.
A version of this story originally appeared in the Good News newsletter. Sign up here!
2025-11-29 20:30:00
NVIDIA股价走势图。| Jonathan Raa/NurPhoto 通过 Getty 图片提供
随着几乎所有东西的价格上涨，而美国工人的工资几乎停滞，像唐纳德·特朗普这样的政界人士试图让我们安心，称经济“表现很好”，股市繁荣。本月早些时候在佛罗里达州的一次活动中，特朗普说：“历史新高，历史新高，历史新高。”
要点总结
As the price of almost everything has increased, and American workers’ wages have all but stalled, politicians like President Donald Trump have tried to ease our minds by telling us that the economy is “doing great” and that the stock market is booming. “Record high, record high, record high,” Trump said at an event earlier this month in Florida.
Still, despite what has been a good year for the stock market, it’s hard to find a day in which a podcaster, influencer, or economist isn’t warning that the AI boom that’s powering the economy could be a bubble — one that is about to burst.
The company that’s driving Wall Street’s positive movement is Nvidia, the most valuable company on the planet. And that’s because the recent rash of data centers popping up across the country are filled with Nvidia’s graphic processing units, or chips.
So why did the health of this single company become an outsized force in the economy? And why does its health scare so many people? Today, Explained co-host Noel King asked economic commentator, educator, and author of In This Economy? Kyla Scanlon.
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.
Recently, the markets have been a rollercoaster. And when you ask why, the answer broadly is because of Nvidia. Why is the world holding its breath for Nvidia? What’s the worry here?
Well, Nvidia is kind of emblematic of the entire AI buildout. So every single tech firm from Microsoft to Meta to Amazon have based all of their future plans around Nvidia. (If you hear anything about “circular financing,” that’s what that means.)
Nvidia is just so wrapped into the broader market — is such a big part of AI — that if they sneeze, everybody else catches a cold. And so markets are a little bit nervous, because the entire AI story, [and] therefore the entire stock market, [and] therefore the entire economy depends on Nvidia maintaining pretty impossible growth metrics.
It really feels like this shouldn’t happen — that there shouldn’t be one company that’s big enough, important enough to make world markets like quiver.
What exactly happened here?
Nvidia just became so big so quickly, and the US economy decided to design itself around AI. You know, 40% of GDP growth is coming from AI buildout. And so Nvidia, because of that concentration, because of the bet that the US economy is making on AI — they have become somewhat of a macro variable.
You can kind of think of their earnings reports like you would a jobs report that we get from the BLS or an inflation report that we get. Earnings day for Nvidia is a test of the AI narrative, and is therefore a test of the US economy. And that just is because we’ve spent so much money on data centers [capital expenditure] — so much money on these chips and these companies just building out continuously. So that’s what happened.
Are there any other companies that hold this sort of sway? Does Walmart or Chevron have that kind of power?
No. Nvidia is such a big part of the S&P 500; it’s almost 8% of the entire index. It’s contributed, I think, a fifth of the index’s total gain this year.
Walmart is not that big of a percentage of the S&P 500, and it has not driven that much growth, that much earnings power, that much investment. Nvidia is really special in that way. …
The S&P 500 has always been pretty top heavy. There’s always been companies that are more important than other companies. But without Nvidia, the story of 2024, 2025, would look like economic stagnation.
You know the old saying, right: The stock market is not the economy. Is Nvidia just playing this enormous role in the markets, or does it represent an outsized portion of other parts of the economy? If Nvidia stumbles, do a million Americans lose their jobs?
I don’t think it would be something that extreme. The stock market is definitely not the economy, but they are increasingly intertwined because the AI narrative is so important. If Nvidia implodes, it wouldn’t be that, like, people who are doctors and bus drivers and construction workers would suddenly be without work.
It would just be that the stock market would collapse, and the economic growth narrative would collapse. And you could see secondary effects. Like maybe the construction firm decides to start laying off people because Nvidia leads to some sort of recession if they do end up imploding. But it would not be a direct correlation, no.
Everybody’s been asking, “Are we in an AI bubble?” And lately I’ve seen people suggesting that Nvidia will be one of the big signs telling us if it’s going to pop.
What do we know about the threat of an AI bubble and where Nvidia plays in?
If I had a nickel for every time somebody talked about the AI bubble, you know, I’d be able to invest in Nvidia. But I think that the way that you can think about it is: Nvidia is the entire AI thesis.
If all of a sudden, Nvidia stumbles — and there’s increasing worries that they’re going to, because their growth path is pretty impressive, and pretty unsustainable because it is so impressive — companies might pull back on spending tens of billions of dollars on data centers. Cloud providers would delay expansion, and startups built around “AI is the future” would face funding problems. The stock market would lose double-digit percentages. The regional construction booms tied to data centers would slow. Places like in Iowa where they’ve helped to revive local economies to a certain extent — everything from steel plants to electrical workers, to construction workers, land developers — would feel the shock.
And then of course if the stock market goes down, ultimately the broad economy does suffer, because then the Federal Reserve would have to come in with some sort of emergency funding plan. President Trump might have to come up with a fiscal policy plan to prevent the bottom from going out and having a massive blow-up.
The worry is if Nvidia does go [down], the entire AI supply chain becomes wobbly. And because the economy and stock market are so tied up into that, it could really lead to some other repercussions.
I wonder, at the end of the day, what you think a company like Nvidia means for the American economy. It is a beast. It takes up a huge share of the market.
What kind of position are we in here that we have a company that is this influential?
Well, Greg Ip from the Wall Street Journal wrote a great piece calling Nvidia the joyless tech revolution. And I think that is a really good way to think about it. The AI trade, if it works, [then] the benefits are going to be accrued to a select few people, right? So companies like Nvidia — people will invest in Nvidia a little bit. Companies like Open AI, companies like Anthropic — they’re going to really benefit if all of this ends up working out.
But the losses from AI are socialized. So if all of a sudden the data centers don’t work, if the AI trade totally blows up, you’re gonna have people’s retirement accounts really suffer, because the S&P 500 is what most people invest in for their retirement account, and Nvidia is a lot of the S&P 500 as we discussed. And then if the data centers don’t work out, you’re going to have a lot of local communities that have pinned hope on these things and have dreamed that they’ll work and add jobs. And so that’s kind of the issue with AI and Nvidia taking up such a big part of the economy.
That’s why Greg is calling it the joyless tech revolution — because a lot of people don’t like this. I think that’s a really important thing to consider. I believe the statistic was 6 out of 10 Americans, essentially, don’t want all of this. They don’t like what the AI companies are promising, especially when the CEOs come on and say that they’re going to take people’s jobs.
Then there’s also a chart from the [Financial Times] that I think encapsulates this broad conversation that we keep having really well, too, where it’s like: AI could either be the end of scarcity, meaning it solves everything; the end of humanity, meaning it kills everybody; or it could add 0.2 percentage points to GDP. And it’s just like how the internet was to a certain extent.
It seems like there’s the potential here that this problem of inequality that we’ve been dealing with now for about a generation could really be exacerbated.
The frustrating thing about the AI conversation is that everybody’s talking about it, but there’s no policy solution yet. We don’t have any idea of how we’re going to re-skill people. We don’t know if we need some form of UBI, universal basic income, to help people out during a time of transition.
We have so many lessons that we could learn from things like what happened to the Rust Belt, when manufacturing went overseas, and how that devastated local communities. We could see something like that happening with AI over time.
2025-11-29 01:00:00
在明尼苏达州奥瓦顿的一家大型工厂农场里，Jennie-O公司饲养火鸡，该公司是美国第二大火鸡生产商，今年还为白宫的火鸡赦免仪式提供火鸡。11月2日晚上，几名动物权益活动人士进入了一间未上锁的谷仓，发现里面火鸡的生存条件极其恶劣。这间谷仓是Jennie-O工厂农场的一部分，而该农场的条件被揭露存在大量动物福利问题。
据活动人士Kecia Doolittle称，他们发现许多火鸡患有严重疾病，甚至有死掉和腐烂的火鸡，还有因互相啄食而受伤的火鸡。此外，一些火鸡因身体过于沉重而无法行走。Doolittle还提到，一些火鸡被束缚，无法获得食物和水。她指出，这些条件违反了明尼苏达州的动物虐待法，该州是少数不将农业活动排除在动物虐待法规之外的州之一。
Jennie-O的母公司Hormel Foods在回应中表示，他们高度重视动物福利，并定期进行设施审核，确保符合国家火鸡协会和美国兽医医学协会的标准。然而，兽医Sherstin Rosenberg认为，Gabriel和Gilbert的状况表明Jennie-O的设施存在严重的动物福利问题。
在火鸡产业中，由于选择性繁殖，火鸡体型变得越来越大，生长速度也加快了。这种做法导致火鸡身体结构不协调，容易出现健康问题。为了防止火鸡互相啄食，农场通常会割掉它们的喙、脚趾和鼻部的肉瘤，而不会给予任何麻醉。此外，火鸡在运输和屠宰过程中并未受到联邦法律的保护。
白宫的火鸡赦免仪式每年都会举行，作为对火鸡产业的宣传。今年，Jennie-O的总裁Steve Lykken负责挑选被赦免的火鸡，它们被送往明尼苏达大学，以确保它们能继续生活。而Doolittle救下的两只火鸡Gabriel和Gilbert则将被送往动物庇护所，过上自由的生活。
尽管火鸡产业的现状令人担忧，但它们仍然是感恩节的象征。这种矛盾的象征意义在白宫的赦免仪式上尤为明显。
Editor’s note: This story was originally published on November 22, 2023, and reflects events that took place that year. We’re republishing it in its original form for this year’s Thanksgiving week.
Late into the night on November 2, a few animal rights activists opened an unlocked barn door and stepped foot into a sea of turkeys living in gruesome conditions. It was one of several barns at a sprawling factory farming operation in Owatonna, Minnesota, that raises turkeys for Jennie-O, the country’s second-largest turkey producer and this year’s supplier to the annual White House turkey pardon ceremony.
“We documented a lot of really horrific health issues,” activist Kecia Doolittle, one of the investigators, told Vox. “It was about as bad as you can imagine.”
They found numerous turkeys who were dead and rotting, Doolittle said, and many who had trouble walking. There were also live birds pecking at dead birds, and dozens of birds with visible wounds — each a sign of cannibalism, a persistent problem in turkey farming.
Doolittle also alleges there were a number of turkeys who were immobilized and unable to access food and water. In a letter to Steele County’s attorney and local law enforcement, Bonnie Klapper — a former assistant US attorney advising Doolittle — said the conditions are a violation of Minnesota’s animal cruelty law, which stipulates that “No person shall deprive any animal over which the person has charge or control of necessary food, water, or shelter.” (Minnesota is one of the few states that don’t exempt agricultural practices from their animal cruelty statute.)
“It smelled terrible,” Doolittle said. The air made her throat burn, likely due to high ammonia levels from the turkeys’ waste, which gives the birds eye and respiratory issues.
The activists found a sign on the property that read, “Jennie-O Turkey Store cares about turkeys — you should, too!”
“Jennie-O Turkey Store takes the welfare of the animals under our care seriously and has robust animal care standards throughout our supply chain,” a spokesperson from Hormel Foods, Jennie-O’s parent company, told Vox via email. “We conduct routine audits at our facilities to ensure that our standards are being met with animal-handling practices and policies set forth by the National Turkey Federation and the American Veterinary Medical Association.”
Doolittle rescued two of the birds — whom she later named Gabriel and Gilbert — and took them to veterinarians in Wisconsin, who urged her to euthanize Gilbert. “They both had really severe infections, they both had parasites,” Doolittle said, but Gilbert was in especially bad shape, with a wound under his wing, an infection on his face, and pecking wounds on part of his genitalia.
But Doolittle wanted to give him a chance to recover. Both birds were treated and given a combination of antibiotic, pain relief, and antiparasitic drugs; Gabriel is on the mend, while Gilbert’s condition remains touch and go.
Sherstin Rosenberg, a veterinarian in California and executive director of a sanctuary for rescued poultry birds, wrote in a veterinary opinion that Gabriel and Gilbert’s condition “suggests serious animal welfare problems” in Jennie-O’s facility.
The findings, while disturbing, are common across the turkey industry. Numerous animal welfare groups have found similar conditions at operations run by Jennie-O’s competitors — even the ones that brand themselves as more humane. That’s because turkey farming is incredibly uniform, with companies using generally the same practices and the same breed — the Broad Breasted White turkey — that’s been bred without regard for their suffering.
Like everything else in the US — cars, homes, cruise ships — the turkey has become supersized.
The poultry industry has made turkeys so big primarily through selective breeding. The Broad Breasted White turkey, which accounts for 99 out of every 100 grocery store turkeys, has been bred to emphasize — you guessed it — the breast, one of the more valuable parts of the bird. These birds grow twice as fast and become nearly twice as big as they did in the 1960s. Being so top-heavy, combined with other health issues caused by rapid growth and the unsanitary factory farming environment, can make it difficult for them to walk.
Another problem arises from their giant breasts: The males get so big that they can’t mount the hens, so they must be bred artificially.
Author Jim Mason detailed this practice in his book The Ethics of What We Eat, co-authored with philosopher Peter Singer. Mason took a job with the turkey giant Butterball to research the book, where, he wrote, he had to hold male turkeys while another worker stimulated them to extract their semen into a syringe using a vacuum pump. Once the syringe was full, it was taken to the henhouse, where Mason would pin hens chest-down while another worker inserted the contents of the syringe into the hen using an air compressor.
Workers at the farm had to do this to one hen every 12 seconds for 10 hours a day. It was “the hardest, fastest, dirtiest, most disgusting, worst-paid work” he had ever done, Mason wrote.
In stressful, crowded environments, turkeys can be aggressive and peck one another, and even commit cannibalism. Instead of giving turkeys more space and better conditions, producers mutilate them to minimize the damage. They cut off a quarter to a third of their beaks, part of their toes, and their snoods — those fleshy protuberances that hang over their beaks — all without pain relief.
Turkeys are excluded from federal laws meant to reduce animal suffering during transport to the slaughterhouse and during slaughter itself, so you can imagine — or see for yourself — how terribly they’re treated in their final hours. According to the nonprofit Animal Welfare Institute, the Jennie-O slaughter plant near the farm Doolittle investigated was cited nine times in 2018 by the US Department of Agriculture for turkeys who’d been mutilated by malfunctioning equipment.
Strangely, despite the horrific reality of turkey farming, we still use the animal as a symbol of giving thanks. Nowhere does the song and dance of celebrating turkeys while we torture them feel more disconcerting than at the White House’s annual turkey pardon.
Every Thanksgiving, the US president “pardons” a turkey or two in what is essentially a PR stunt for the turkey industry, as the birds are selected by the chair of the National Turkey Federation, an industry trade association. This year, that was Steve Lykken, president of Jennie-O.
The two turkeys selected for this year’s pardon — named Liberty and Bell — could have ended up among the 46 million or so birds on Thanksgiving tables this year. Instead, they were transported from Minnesota, the country’s top turkey-producing state, to Washington, DC, in a stretch black Cadillac Escalade. “They’re on their way in a pretty lavish coach,” Lykken told Minnesota Public Radio.
The annual story makes for feel-good if hammy coverage by the nation’s largest news organizations, but it papers over the darkness of American factory farming — including not just the animal cruelty but also the dangerous working conditions at slaughterhouses, environmental pollution, and unfair treatment of turkey contract farmers.
The White House did not respond to a request for comment about the Jennie-O investigation video.
This year, industry is especially looking forward to the pardon amid the devastating bird flu. The disease, which has been resurging this fall, has resulted in the killing of 11.5 million potentially infected turkeys since early 2022. Increasingly, producers are killing the birds in the most brutal fashion imaginable, deploying a method called “ventilation shutdown plus” that uses industrial heaters to kill them via heatstroke over the course of hours.
“To have something that’s fun, that can draw positive attention to our industry, is very welcomed” in light of the outbreak, Ashley Kohls, executive director of the Minnesota Turkey Growers Association, told Minnesota Public Radio about this year’s pardon.
This week, Liberty and Bell will be moved to the University of Minnesota to live out the rest of their lives. If the turkeys knew what went on there, they might not want to go: The university helped build the state’s turkey industry and still conducts research on turkeys to ensure the industry’s success. The university’s interim president formerly served as the president of Jennie-O and the CEO of Hormel, its parent company.
Meanwhile, Doolittle’s pardoned turkeys, Gabriel and Gilbert, assuming both survive, will spend the rest of their lives at an animal sanctuary, showing humans what these birds can be like when allowed to live on their own terms. “They’re just the most curious, loving, intelligent guys,” Doolittle said.
A version of this story originally appeared in the Future Perfect newsletter. Sign up here!
2025-11-28 20:30:00
在经济不确定的时期，Z世代正在通过购买价格实惠的“身份象征”商品来获取社交媒体的关注度并脱颖而出。例如，星巴克最近推出了一款以熊为造型的限量版杯子“Bearista”，原价30美元，最初在韩国销售，但很快在美国的星巴克门店售罄。然而，在抢购过程中，一些顾客甚至与店员发生争执，甚至互相推搡。这些抢购行为在社交媒体上引发热议，甚至登上TikTok热搜。星巴克对此表示歉意，但并未确认是否会重新补货。目前，这款杯子在eBay上以高价出售。
与此同时，成功购得“Bearista”杯的顾客纷纷在社交媒体上分享自己的“胜利”。这些杯子只是今年引发热潮的众多“看似随机但难以获得”的商品之一，例如Labubus、Owala水瓶和Trader Joe’s迷你帆布包等。虽然低价小商品引发抢购的现象并不新鲜（如Cabbage Patch Kids和Beanie Babies），但如今这些商品已成为一种新的身份象征，成为年轻人展示个性和独特性的创新方式。
传统上，身份象征被认为是昂贵的奢侈品，但如今在经济压力日益增长的背景下，年轻人开始寻找其他方式来体现自己的社会地位。即使是一些价格不高的商品，只要能获得，就能成为一种“炫耀”。即使是富裕人群，也开始接受这些低调但有“圈内”意味的象征。这些商品的稀缺性使得购买行为本身成为一种社交体验，甚至在社交媒体上形成了一种独特的“寻宝”文化。例如，有人会拍摄自己在超市或大型卖场中寻找限量版Stanley保温杯或迪拜巧克力的过程，甚至有人会去翻垃圾桶寻找商品。这些行为虽然可能引发冲突或盗窃，但却为产品带来了更多关注。
此外，这些商品的流行也反映了Z世代对“可负担的奢华”（affordable luxuries）的追求，这被一些人视为经济衰退的征兆。这种现象类似于2001年经济危机期间的“口红效应”（lipstick effect），即消费者在经济困难时更倾向于购买价格低廉但能带来愉悦感的商品。然而，哥伦比亚大学商学院副教授Silvia Bellezza认为，Z世代对这些小商品的痴迷并不完全是因为经济压力，而可能更多是出于对“品味”或“文化资本”的追求。她指出，传统的高端象征（如豪车和名牌服装）在大规模生产后变得过于普遍，因此消费者，包括富裕人群，开始寻找更独特的方式来彰显自己的身份。
“将高端与低端结合是一种聪明的方式，可以让你比那些只追求传统高端商品的人更具个性，还能掌控自己的时尚风格。”Bellezza表示。这种趋势也解释了为什么名人和亿万富翁们也开始佩戴Labubus或使用帆布包。此外，消费者希望拥有只有特定群体才能理解或识别的商品，这种行为被称为“水平信号”（horizontal signaling）。例如，Trader Joe’s的帆布包不仅是一个实用的购物袋，更是一种潮流符号。
由于社交媒体在Z世代购物行为中扮演着重要角色，拥有这些“身份象征”更多是为了获得社交影响力，而非实际的财富展示。尽管抢购这些商品可能带来诸多不便，但对于一个经济前景不明朗的群体来说，这种行为似乎合情合理。在这样的背景下，抢购行为不仅是一种消费方式，更是一种社交资本的积累。
A cuddly animal wearing a beanie should not incite violence. And yet, this is what occurred in some Starbucks shops earlier this month after the coffee chain released a limited line of teddy bear-shaped tumblers as a part of their holiday merchandise.
The “Bearista” cup, which was originally sold in South Korea and retails for $30, is reportedly sold out across Starbucks locations in the United States — but not before customers hassled baristas and tussled with each other in line to get their hands on one. Stories of frustrated Starbucks fans standing in long lines at the crack of dawn and accusing employees of hogging the cups made headlines and went viral on TikTok. Starbucks apologized to customers for the frenzy but stopped short of confirming whether they would restock the cups.
For now, these glass critters are being sold on eBay, sometimes for exorbitant prices. Meanwhile, shoppers who managed to procure the Bearista cup in store are posting their wins.
The Bearista cup is just the latest entry on a list of seemingly random but extremely hard-to-get merchandise that’s gone viral this year, from Labubus to Owala water bottles to Trader Joe’s micro tote bags. That certain low-cost novelty items can attract huge lines isn’t exactly new. (Cabbage Patch Kids and Beanie Babies are canonical examples.) What’s new is the extent to which these relatively cheap trinkets have become status symbols and an innovative way to distinguish oneself.
Status symbols are traditionally thought of as expensive luxury items that indicate one’s class. But in a time when affordability is driving the political conversation, young consumers are finding alternative ways to convey their social stature. It isn’t a flex anymore to splurge on a designer bag or an expensive car. Status can be earned by getting your hands on something rare, even if it only costs 30 bucks. Even wealthy people are embracing these less glitzy but in-the-know signifiers.
In the case of all these conspicuous or wearable commodities, the goal is to invest and partake in the online attention economy.
While some of the hottest items of the year have necessitated elaborate hunts or long wait times, this isn’t always intentional on the brand’s part. Many companies use artificial scarcity as a marketing tactic, partaking in seasonal or limited product drops to boost sales, while others may not anticipate the initial demand of new products and run out of supply, according to Tara Sinclair, head of the economics department at George Washington University. In any case, these shortages lend themselves to social media content that sparks intrigue and turns the purchasing of these products into full-blown experiences.
On TikTok and YouTube, the search to find a special-edition Stanley cup or a Dubai chocolate bar has become its own subgenre. Users post trips to their local grocers or big box stores where they scour the aisles, either successfully or unsuccessfully, for the latest trendy purchase. Others have filmed themselves going dumpster diving behind stores. In some cases, recorded confrontations between especially committed customers, and even theft of these items, have helped these products generate buzz. Much of the conversation around Stanley cups and Labubus, for example, was about the outsized emotional reactions people seem to have to them, adding a mysterious value to these products.
Given that these products are not financially out of reach for working and middle-class consumers, the thrill of attaining these items seems largely wrapped up in beating your competition, whether through time, dedication, or pure luck.
That said, these items still signal a level of financial privilege. (A $30 mug is more expensive than the paper cup that you usually get with your coffee at Starbucks.) And the amount of time and energy spent scavenging these items “is a way of signaling being part of a club,” says Sinclair.
“When we think about products of scarcity, there’s typically two ways that we pay for them, money or our time,” she says. “Spending time scouring a store is not so different from spending your money on it because you could, otherwise, potentially be working and earning money during that time.”
Even so, it’s notable that our idea of status symbols is expanding to include a variety of low-end and easily collectible things. For a generation riddled with economic anxiety and navigating a slimming job market, it feels appropriate that cost-friendly items are suddenly carrying a lot more esteem.
This fits into a larger trend of Gen Z spending on so-called “affordable luxuries,” that some have theorized as a recession indicator. To many, the current craze around these low-cost products seems like a re-treading of the “lipstick effect,” a term that dates back to the 2001 recession. At the time, Estée Lauder chair Leonard Lauder noted that the company had experienced a rise in lipstick sales, inspiring a theory that says consumers are less likely to spend on expensive, luxury goods in the midst of an economic crisis, instead opting for cheap indulgences. After all, luxury spending is slowing down and could continue this trajectory for a while.
But Silvia Bellezza, associate professor of business at Columbia University, isn’t totally convinced that Gen Z’s obsession with trinkets and tumblers is primarily a response to harsh economic times, given that people are always falling for these novelty items. These alternative status symbols could also be driven by an interest in acquiring taste or “cultural capital.”
Bellezza has found that “traditional markers of superiority,” like cars and designer clothes, become too mainstream and diluted when they are mass-produced. This leads to consumers, including affluent people, finding more clever and original ways to distinguish themselves.
“Mixing and matching high and low is a very clever way to stand out and show that you’re even superior to engaging in all the traditional high-status products, and you can dictate your own fashion,” says Bellezza.
This would probably explain why celebrities and billionaires alike are incorporating Lababus and canvas totes into their high fashion wear. Plus, she says consumers want to be seen with items that can only be perceived and understood by a certain subset of people, a behavior referred to as “horizontal signaling.” A Trader Joe’s tote, for example, is mostly understood as a trendy fashion item and not just a utility bag to a specific group of young, very online urbanites.
Given that social media is such an integral part of Gen Z’s shopping experiences, it’s only natural that owning status symbols is more about achieving social clout than signaling actual currency. Waiting in line, fighting crowds, and tracking down these affordable items may be extremely inconvenient, but it’s suitable for a cohort with an uncertain financial future. In the meantime, why not rack up views?
2025-11-28 20:00:00
2025年11月19日，美国司法部长帕姆·邦迪在华盛顿特区的司法部新闻发布会上亮相。| 埃里克·李/彭博社 通过盖蒂图片社提供
唐纳德·特朗普试图将司法部当作自己的私人律师事务所。在他的领导下，司法部会因个人政治原因而放弃某些案件，或在缺乏证据的情况下启动案件。同时，司法部还试图起诉特朗普的对手和政敌，包括前FBI局长詹姆斯·科米和纽约州检察长利蒂西亚·詹姆斯，后者在2022年曾对特朗普提起民事诉讼。这些案件遭遇了一些挑战：上周一，一名联邦法官驳回了政府对科米和詹姆斯的指控。但特朗普试图利用司法部进行政治操作的做法，也在司法部内部留下了深远的影响。
《纽约时报》杂志的撰稿人艾米丽·巴泽伦采访了自2023年1月以来辞职或被解雇的数千名司法部律师。通过他们的故事，她带领我们了解司法部内部的动荡、对特朗普指令的抵制，以及这一切将如何影响未来。
以下是采访内容的节选，已进行删减和润色。完整播客内容更丰富，欢迎收听《Today, Explained》。
让我们回到事情的开始。在特朗普新政府的第一天，司法部发生了什么？
在第一天，特朗普就明确表示，那些对他个人忠诚的律师将负责司法部的工作。这首先体现在司法部长帕姆·邦迪身上，但还有其他人员被他安排上任。此外，他还赦免了在1月6日美国国会骚乱中被指控参与暴乱和暴力的人。这是一项司法部历史上最大的调查。检察官们投入大量精力处理这些案件，但特朗普的赦免决定对这些工作人员来说是一个巨大的打击。
2月，特朗普提名的司法部长帕姆·邦迪获得确认。请介绍一下她以及所谓的“帕姆·邦迪混音带”事件。
帕姆·邦迪是佛罗里达州前检察官，有丰富的检控经验，也曾担任特朗普的私人律师。有律师提到，她上任第一天就发布了14份备忘录，暂停了检察官们一直重视并优先处理的某些腐败相关法律的执行。她强调“积极辩护”，即律师的忠诚应是对总统而非宪法。这些举措表明司法部的优先事项正在发生转变。
这些优先事项转向了什么？
转向了特朗普的议程：反对任何多样性努力，将资源转向移民事务，而远离司法部传统职责，如起诉公共腐败。
我想问问你，这些律师在采访中提到的一些具体案例。例如，有律师说他们因为梅尔·吉布森而丢了工作，这是怎么回事？
这是我的同事拉切尔·波瑟采访的赦免律师伊丽莎白·奥耶的故事。她被重新分配到一个专门调查枪支犯罪赦免的部门。有人建议她赦免梅尔·吉布森，他因家庭暴力被判处轻罪。奥耶被告知，吉布森与总统有私人关系，因此她被要求寻找理由赦免他。但考虑到他有家庭暴力的记录，她非常不愿意这样做，最终拒绝了这一要求，随即被解雇。
接下来，我们谈谈司法部民事权利部门的情况。你提到过一个叫做“消防员案件”的故事，这是怎么回事？
我的同事拉切尔采访了一位名叫布莱恩·麦金太尔的律师，他在佐治亚州处理一个案件，发现黑人和白人在申请消防员职位时比例相近，但90%的录用都是白人。当他们询问消防部门原因时，被告知黑人申请人通常有更多学生贷款债务。于是消防部门说：“如果一个消防员深陷债务，他在救火时可能会偷走奶奶的珍珠。”因此，他们减少了对黑人申请人的录用。民事权利部门随后提起诉讼。2月，他们收到通知，要求撤回案件，并要求在撤回文件中加入“这是关于反向歧视”的措辞，即真正的受害者是白人。这些律师对此感到非常困扰，因为他们认为这并不属实，最终拒绝签署相关文件。
随着年份的推进，特朗普政府如何在司法部内部分配资源？他优先处理哪些事务？
有一项重要命令指出，大约三分之一的执法人员和资源将被转移到移民事务上。这意味着FBI特工将不再从事他们以前的工作，如白领犯罪、国家安全、反恐和儿童色情等案件。这些案件通常需要大量人力和时间。因此，如果FBI特工被派去抓移民，他们将无法处理这些更长期的重要案件，而这些案件对检察官们来说，对于保障美国人的安全至关重要。
到了9月下旬，特朗普要求司法部向他支付2.3亿美元，作为对其在拜登政府期间调查的补偿。这一要求在司法部内部如何被处理？
这是一项前所未有的要求。而且，决定特朗普是否能获得这笔巨额赔偿的人，正是他的任命者，包括司法部长帕姆·邦迪及其副手托德·布兰奇。从我们采访的司法部律师角度来看，这显得非常荒谬和腐败。他们无法理解特朗普为何认为这是联邦资金的适当用途。
其中一位消息人士表示，不报复是下一届政府的一大考验。他说，他脑海中有一份名单，列出了那些帮助特朗普政府的人，他希望在下一届政府中追究他们的责任。你从这次报道中是否感到，这种报复循环可能正在形成？
目前还不能确定是否会形成这种循环，但确实存在很大的诱惑。因为有些人会觉得自己被那些他们曾目睹不道德或背叛行为的人包围，很难放下这些情绪。这可能会引发就业后果，例如是否所有人都能保住职位，甚至是否会有刑事调查。这种报复行为可能会严重扰乱司法体系。
另一位消息人士表示，普通美国人并不太关心司法部发生的事情，因为他们觉得这与自己无关。是否有一种观点认为这实际上影响到了我们？
我认为确实如此：法律的权威。法律的稳定性对于美国的繁荣和社会福祉至关重要。实际上，法律最重要的作用就是提供稳定性。当总统可以随意动用联邦执法力量对付任何人时，人们就会觉得这可能发生在自己身上。但与过去相比，这种报复的可能性已经不同了。自水门事件以来，美国一直有意维持白宫与司法部之间明确的界限，以防止政治影响渗透到调查和起诉中。一旦这种界限被打破，最终会在美国人的生活中以各种方式显现出来，即使它最初似乎只涉及像詹姆斯·科米和利蒂西亚·詹姆斯这样的人。
President Donald Trump has been trying to use the Department of Justice as his personal law firm. Under Trump’s DOJ, cases are dropped for personal political reasons or built without evidence. The DOJ has also sought to prosecute Trump’s adversaries and political foes, including James Comey, the former FBI director, and Letitia James, the New York attorney general whose office filed a civil lawsuit against Trump in 2022.
Those cases have faced some challenges: On Monday, a federal judge threw out the government’s charges against Comey and James.
But Trump’s attempts to use the Justice Department for political ends are leaving their mark inside the department as well. Emily Bazelon, a staff writer at the New York Times Magazine, spoke to some of the thousands of DOJ attorneys who have resigned or been fired since January. Through their stories, she navigated us around the turmoil happening at the department, the pushback to Trump’s directives, and where it all leaves us.
Below is an excerpt of the 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.
Let’s go back to the beginning. It is day one of the new Trump administration, and what is going on at the DOJ?
On the very first day, Trump first of all makes it clear that lawyers who are personally loyal to him are going to be in charge of the Justice Department. That starts with the attorney general, Pam Bondi, but there are other people he puts in place as well. And then the other thing he did was that he pardoned all of the people accused of rioting and violence on January 6th in the insurrection at the US Capitol.
This was the biggest investigation in the history of the Justice Department. They felt really strongly that this was a really important signal to send – that the US government would not tolerate the kind of violence and disruption that could have derailed the peaceful transfer of power after the 2020 election. Prosecutors had devoted themselves to these cases. It was just a huge blow to the people who worked on all of these matters.
In February, President Trump’s pick for attorney general, Pam Bondi, is confirmed. Tell us about her and tell us what the so-called “Pam Bondi mixtape” is.
Pam Bondi is a former state’s attorney from Florida. She had prosecutorial experience. She was also a personal lawyer for Donald Trump. And this kind of idea of a mixtape, as one of the lawyers put to us, was that she issued a flurry of 14 memos on her first day. She paused the enforcement of certain corruption laws that prosecutors traditionally work hard on and make a priority. She talked about zealous advocacy — the idea of the lawyer’s commitment as being a commitment that was to the President as opposed to simply the Constitution. And there were other kinds of moves like that, that just made it clear that all of the priorities of the Justice Department were shifting.
What were they shifting to?
They were shifting to President Trump’s agenda: an agenda that was against any kind of diversity efforts, an agenda that was toward immigration work and away from traditional aspects of the Justice Department’s purview, like prosecuting public corruption.
I want to ask you about some of the specific cases that these attorneys talked to you about when you interviewed them. There was a lawyer who said they lost their job in April because of Mel Gibson. What happened there?
This is the pardon attorney Elizabeth Oyer, who my colleague Rachel Poser interviewed. And Oyer’s story is that she was reassigned to this unit that was looking into pardons of people who’d been accused of gun crimes. And the idea came up of pardoning Mel Gibson, who had a misdemeanor conviction for domestic violence.
And it’s been strongly suggested to her, she says, that Gibson had a personal relationship with the president. She was basically given the message that she needed to find some way to pardon Mel Gibson. But because of his history of domestic violence, she was very reluctant to do that. She said no to the idea of a pardon, and she was immediately fired.
Moving forward in time, let’s talk about what you learned was happening in the Civil Rights Division. There’s one story you tell about something called the “Firefighter Cases.” What happened there?
My colleague Rachel interviewed a lawyer named Brian McIntyre in the Civil Rights Division. He had been working on a case in Georgia where Black people and white people were applying for positions in the fire department at about the same rate, but 90 percent of the hires were white people. And so Brian McIntyre was wondering why.
And when they asked the fire department, the answer was that Black people tended to have more student loan debt. And so then the fire department said, “Okay, well, our problem with that is if you have a firefighter and he’s deeply in debt and fighting a fire, he might steal grandma’s pearls.” So this was apparently the reason for hiring fewer Black firefighters. And the Civil Rights Division sued.
In February, they got a note saying that the Attorney General Pam Bondi wanted to withdraw the case and they went further in a way that was really distressing to the lawyers by asking for additional language in dismissing the case that would say that it was all about reverse discrimination. In other words, the real victims here were white people. And so these lawyers in the Civil Rights division, they really wrestled with whether they could sign this order because they didn’t think it was true. And in the end they did not sign it.
As the year progresses, how does the Trump administration start divvying up resources at the DOJ? What do we see Trump prioritizing?
There’s a really important order that happens where about a third of the manpower and resources of law enforcement agents is supposed to start going to immigration work. And that means that these FBI agents are not going to be doing the things they were doing before because their work hours are a finite resource.
Prosecutors told us that they saw these agents being pulled off of cases involving white collar crime or national security, counter-terrorism, child exploitation. Those are the kinds of big cases that just take a lot of labor. And so if you have your FBI agents out on the street picking up people for immigration detention, then they’re not going to be able to do these more longer-term cases that, in the view of the prosecutors, are very important for keeping Americans safe.
Moving forward to late September, Donald Trump has demanded that the DOJ pay him $230 million for investigations into him that happened during the Biden administration. How does that play out within his Department of Justice?
This is a really unprecedented demand. And also remember that the people who are going to decide whether Trump gets this big payout are his appointees, his former lawyers in the Justice Department, right? Pam Bondi and her deputy, Todd Blanche.
From the point of view of the Justice Department lawyers we interviewed, this just seemed comically corrupt to them. They just really couldn’t imagine how the president could think this was an appropriate use of federal funds.
One of your sources told you it would take a lot of restraint not to retaliate in the next administration. This person said they have a list in their head of career people who are helping the administration they want to hold to account. Did you come away from this reporting concerned that there is a cycle of retribution here that may be becoming entrenched?
It’s too soon to say there is going to be a lot of temptation to move in that direction because some people are going to feel like they’re surrounded by people who they watched do things that were unethical or traitorous to the colleagues around them. It’s hard to let all of that go.
I think there are different ways that could be addressed. There are employment repercussions, like questions of whether everyone gets to stay in the job. And then there’s the much more serious question of whether they’re going to be criminal investigations. That’s the kind of tit for tat retaliation that I think could really send the justice system into a tailspin.
Another of your sources tells you that the average American does not really care what is happening at the Justice Department because we think it doesn’t affect us. Is there an argument that this does in fact affect us, that we should really care what’s going on here?
I think there is: the rule of law. The idea of the stability of law is vital to American prosperity and social well-being, right? I mean, stability is honestly the most important thing we get from law. And when you live in a country where the president can turn the huge might of federal law enforcement against anyone he wants, then you’re kind of betting it’s not going to be you. But the odds are not the same as they were before when this kind of retribution was just off the table.
And since Watergate, we have lived in a country where there was a very deliberate, carefully erected separation between the White House and its political influence and investigations and criminal prosecutions from the Justice Department. So once that is gone, eventually you see that play out in all kinds of ways in Americans’ lives. Even if it starts by seeming it’s just about a few people like James Comey and Letitia James.
2025-11-28 19:45:00
最近的就业数据描绘了劳动力市场严峻的图景，也显示出人工智能（AI）正在对市场造成显著影响。今年早些时候，人们已经对毕业生失业问题发出警告，而最新的报告显示，AI的影响正在扩大到更广泛的工人群体。10月份，超过15万名员工被裁员，这是过去二十年中最糟糕的十月，其中约有5万名裁员与AI有关。2025年至今，裁员数量已超过2020年以来的任何一年。尽管公司公开将AI视为裁员的原因，但目前还无法确定AI到底在多大程度上导致了这些失业。由耶鲁预算实验室和布鲁金斯学会的研究团队认为，AI对整体劳动力市场的影响并不比互联网或个人电脑更大，而近期大学毕业生的失业更多是由于行业特定因素造成的。然而，Anthropic的首席执行官Dario Amodei却预测，AI可能会消除一半的初级白领工作。那么，到底哪种情况更准确呢？我们对AI的总体影响还有很多未知，比如AI泡沫的问题，目前还无法判断AI是否能实现其最雄心勃勃的承诺，或者是否会比以往的技术革命更具变革性。为了更具体地探讨就业问题，我联系了麻省理工学院计算机科学与人工智能实验室（CSAIL）的首席研究科学家Neil Thompson。他一直在研究自动化如何影响劳动力价值，以及为什么前沿模型的边际回报会塑造AI的未来。过去几年，他的研究一直在反驳“自动化总是对工人不利”的观点，并指出AI不会夺走所有人的工作。但最近几个月，我们看到大量因AI而失业的案例。他推测，这可能是因为同时存在两种现象：一方面，AI在经济中变得越来越普遍，确实会取代一些工作，比如客服岗位；另一方面，这些系统可能无法完成像失业数据所暗示的那么多任务，因此，一些裁员可能是公司主动削减岗位并归咎于AI，或者提前裁员以推动AI应用。此外，许多公司虽然声称AI能做更多，但实际应用中仍有许多“最后一公里”的成本问题，比如需要特定数据支持，这会阻碍AI的普及。除了成本问题，还有“好”与“足够好”的区别。AI可能在某些任务上表现优异，但未必能完全取代人类。在2024年，我和MIT的同事David Autor发表了一篇论文，用“专业技能”框架来理解自动化如何影响劳动力价值。历史上，自动化并不总是带来负面影响。例如，过去40年的计算机化虽然自动化了很多常规任务，但这些任务的工资并没有下降，有些甚至上升了。关键在于被自动化的是哪些任务。如果自动化的是高技能任务，那么工资可能下降，但从业者数量会增加；如果自动化的是低技能任务，那么工资可能上升，但从业者数量会减少。例如，出租车司机的高技能任务（如熟悉城市道路）被Google Maps等工具取代，导致他们的工资下降，但更多人可以从事类似的工作。而校对员的低技能任务（如拼写检查）被自动化，他们的工资上升，但人数减少。AI并不是第一个在计算机时代改变工作的技术，但是否适用同样的“专业技能”框架呢？比如在工业革命时期，自动化纺织工人的工作是否也出现了类似模式？我的合著者曾提到，熟练工匠（如轮匠和铁匠）在工业化过程中被生产线上平均技能较低的工人取代，但生产效率和规模大幅提高。同样，随着AI的发展，我们可能会在某些领域变得更加专业，而不再做那些基础工作。谷歌和OpenAI等公司承诺其技术将超越自动化基本任务，他们投入数百亿美元建设基础设施，以实现更高级的AI。然而，我们最近听到很多关于AI泡沫的讨论，因为尚不清楚这些工具是否能在实际应用之前证明其价值。那么，我们如何知道AI是否已经证明了自己？我认为，AI是否能证明自己并不是问题，因为其能力正在迅速提升，未来将非常有用，并带来大量采用和好处。问题在于估值是否合理。AI的实用性是确定的，但其价值是否被高估？这将影响我们如何调整。最近皮尤研究中心的一项调查显示，美国人对AI技术更多是担忧而非兴奋。为什么AI如此不受欢迎？我认为，人们担心AI会改变他们的工作，甚至取代他们，这种焦虑是可以理解的。因为AI是一个非常强大的工具，它可能会改变很多人的工作，包括我的。面对这种不确定性，人们很难知道自己的工作会受到多大影响，或者需要做出多大的调整，这可能会带来痛苦。不过，我们也会在未来一段时间内了解更多。另一个挑战是，我们无法提前预知新技术会带来哪些新任务。但历史表明，新技术往往会创造新的任务和工作机会。因此，我们应该对未来充满信心，相信会有新的岗位出现。尽管如此，这种转变可能需要时间，如果转变过程过于迅速，可能会给经济带来更大的冲击。听起来你是在说，我们对未知的恐惧是当前的主要问题，而我们确实面临许多未知。但历史上我们已经经历过重大技术变革，只是不知道这次需要多长时间，或者我们最终会从事什么工作。这并不令人感到安心。不过，我想补充一点：从历史来看，新技术的出现确实会带来一些经济波动，有些人会受到伤害，但我们也应该意识到这一点，并做好准备。在中长期来看，我们通常能够很好地适应。就AI而言，我认为我们可以从历史经验中获得一些安慰。问题是，AI是否在某些方面不同于以往的技术，从而导致不同的结果？我认为，那些认为AI会迅速发展到通用人工智能（AGI）或超级智能的人会认为是的。如果AI能在短时间内完成所有人类能做的事，那么这将与以往的技术革命不同，调整起来会更加困难。但如果AI的发展是渐进的，逐步完成某些任务，而其他任务则需要较长时间，那么我们就有机会像过去一样进行适应。这篇文章的版本也发表在User Friendly通讯中。点击此处订阅，以免错过下一篇文章！
The most recent jobs numbers paint a pretty grim picture of the labor market and the apparent havoc AI is wreaking on it. After warnings about unemployment among recent grads earlier this year, the newest report suggests that AI’s impact is reaching a broader group of workers. There were over 150,000 layoffs in October, which makes it the worst October for layoffs in over two decades, and about 50,000 of those have been attributed to AI. Overall, 2025 has seen more job cuts than any year since 2020.
It’s too soon to tell how much AI is really to blame for these job losses, even if companies are blaming AI in public statements. A team of researchers from the Yale Budget Lab and Brookings has argued that the broader labor market isn’t being disrupted any more by AI than it was by the internet or PCs, and that recent college grads are being displaced due to sector-specific factors. Anthropic CEO Dario Amodei, however, has predicted that AI could eliminate half of entry-level white collar jobs. So, which is it?
There is a lot we don’t know about what will happen with AI in general — looking at you, AI bubble — and it’s too soon to tell whether AI will actually deliver on its most ambitious promises or be more transformative than past tech revolutions.
But, to shed some light on the jobs question in particular, I called up Neil Thompson, principal research scientist at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL). He’s been studying everything from why diminishing returns on frontier models will shape AI’s future to how automation changes the value of labor. Our conversation has been edited for length and clarity.
For the past couple of years, your work has pushed back on the idea that automation is always bad for workers and that AI will take all of our jobs. But, in the past few months, we’ve seen tens of thousands of job losses attributed to AI. What’s going on?
My guess is that we have two different phenomena going on at the same time. One is that AI is becoming more prevalent in the economy. I think, for some cases, like customer service, that’s probably pretty legitimate. Indeed, these systems seem awfully good at those tasks, and so, there are going to be some jobs that are being taken over by these systems.
At the same time, it would be surprising to me if these systems were able to do as many things as the job loss numbers imply. And so, I suspect that there’s also a mix of either people deciding to cut the jobs and put some of that blame on AI, or they’re cutting the jobs in advance with an aim to do more AI. They’re sort of pushing their businesses towards it and seeing what’s going to happen.
Why is there such dissonance between those who say AI will take away half our jobs and those who say AI isn’t the reason we’re seeing so much upheaval in the labor market?
A whole bunch of people are talking about incredibly rapid change — a capability increase, which could do things that humans can do. For most businesses there are very large last-mile costs that are involved with actually adopting these systems. Someone using ChatGPT just in the interface is very different than “we now run our business and trust that every time the system is going to run, it’s going to get it right.” That’s a different level. You often need to bring in specific data. There are a lot of costs that come with that. So, these last-mile costs can be very important and can really slow adoption even when systems are quite good.
Apart from that cost, there’s also a matter of a system being good, and a system being good enough to be better than a human. They’re not quite the same thing.
Earlier this year, you published a paper with your MIT colleague David Autor that used expertise as a framework for understanding how automation affects the value of labor. Historically, it’s not all bad, right?
When we think of automation, we have in our mind a sort of doom scenario, where, as automation happens, the number of jobs that are out there in that occupation go down, the wages in that occupation go down, and you’re like, “boy, this has been a pretty terrible story.”
But, if you look at the last 40 years of automation — this is not AI automation, this is just computerization and things like that — we know that a lot of routine tasks were automated by this process. If you look at people who had routine tasks, what you find is a bunch of that stuff got automated, but also their wages didn’t go down. Some went up, some went down. That’s kind of a puzzle.
What we think is going on is that, when automation happens to a particular occupation, it really, really matters which of the tasks of that occupation are getting automated. In particular, if you have automation of high-expert tasks — so the things that you do that are most expert — that has one effect, and if you have automation at the least-expert tasks, you’ll get a different effect.
Can you give me a couple of examples?
Think about taxi drivers. The most expert thing you did was know all of the roads in a city. You knew all the little back roads. You knew all the little shortcuts. You were the expert on that. Then, Google Maps and MapQuest come in, and all of a sudden, anybody who can drive a car can do a pretty good job of doing that. In that case, your most expert tasks got automated away. Because the most expert things are gone, your wages go down.
But, counter to this doom cycle version of this, wages go down, but the number of people in that profession goes up, because now, a whole bunch of people who didn’t used to know all the streets can suddenly drive an Uber.
At the other extreme, think of proofreaders. Spellcheck comes in. A whole bunch of stuff that they used to do is now automated, but it was the least expert thing that they did. The meaningful thing they did was to reorganize your paragraphs and make sure that you were thinking about the right thing and phrasing things in the right way, not the spelling part.
So, if you look at what happens to them, their least expert tasks got automated. What was left was more expert. And so, because they were using their expert stuff more of the time, their wages have actually gone up faster than the average — but there are now fewer of them.
So, you have this interesting effect where the Uber drivers’ wages went down, but there were more of them. And for the proofreaders, wages went up, and there were fewer of them. And both of those have pluses and minuses.
So, clearly, AI is not the first technology to automate aspects of work in the computer era. But does the same expertise framework hold true further back in history? Would we see similar patterns in the Industrial Revolution and automating textile workers’ work?
One of the examples that my co-author likes to talk about is skilled artisans. Think about the wheelwrights, and the blacksmith, and all of those people, these used to be incredibly expert jobs. And through industrialization, we figured out how to do that on production lines and other places where the average expertise was lower, but there were vastly more wheels being produced and vastly more people involved in the production of wheels.
And then, of course, we have lots of modern examples as automation comes in, and some of the things that we do get automated, we actually become more expert in the things we’re doing because we don’t have to do the basic things anymore.
Companies like Google and OpenAI are promising that their technology will do much more than automate basic tasks, and they’re spending hundreds of billions of dollars on infrastructure to make it — call it artificial general intelligence or superintelligence — happen. We’re hearing a lot about an AI bubble lately, because it’s not clear if these tools will actually work before the bill comes due. How will we know when AI has proven itself?
I don’t think that the question is really, is AI going to prove itself. I think it is clear that these capabilities are improving fast enough. It’s going to be incredibly useful, I think, and I think there’s going to be a lot of adoption. There’s going to be a lot of benefits that flow from it.
To me, the question in terms of the AI bubble is more about valuations. This is going to be useful, but is that the right valuation? It is going to matter a lot. It’s going to have a lot of these effects. The question is, are we building out even faster than those effects are going to kick in, or the opposite?
A recent Pew Research Center survey showed that Americans are more concerned than excited about the technology. Why is AI so unpopular?
I want to be hesitant about putting myself too much in people’s heads, but I think it is understandable that people have anxiety about what AI is going to do and how it’s going to change their jobs, because it’s a very powerful tool. I think it will change a lot of people’s jobs — yours included, mine included.
I think it is particularly hard when faced with that and not knowing how much of the job is going to be replaced or how much am I going to have to adjust in ways that could be painful. I think we will learn more about that in the next little while.
There’s a second piece which is really, really hard. Historically, when new technologies have come in and automated things, humans have moved to doing new tasks. New tasks are created that didn’t exist before but are actually important for employment. We really don’t know what those new tasks are going to be ahead of time. That lack of visibility is a challenge. But it is worth saying that, historically, there’s been a remarkable wellspring of new tasks and new jobs that have emerged. And so, I think we should feel confident that there are going to be a bunch of those that will come.
There will be a transition. In many cases, we should think of that as being similar to previous transformations. The question is how fast it happens. If it’s medium- to long-term, humans are pretty good at saying, “okay, if these are new tasks that we are particularly good at and the technology is not, let’s adapt to do those tasks.” But if it happens all at once, and a lot of the transitions and displacement happens in a compressed period of time, that’s going to make it much harder for the economy to adjust.
It sounds like you’re saying that there’s a fear of the unknown, and there are a lot of unknowns right now. But, we’ve gone through major technological transformations before this one. We just don’t know how long it will take, or what we’ll be doing on the other side of it. That doesn’t sound super comforting.
Let me just add a little twist to that. It is definitely the case that if you look historically, we have seen patterns where new technologies come in. There is some churn in the economy, some people are hurt by that, and we should be cognizant of that. We should expect that could happen now, as well. But in the medium term, we adjust well.
In terms of AI, I think we can take some comfort from those historical lessons. And the question is just: Is AI in some way different than these previous technologies that would make us think that we would get a different outcome?
I think the people who think that we’re going to get to AGI quickly, their answer would be yes. If it can do everything we can do, and it can do that next year or the year after, that is very different than previous technologies. That makes it pretty hard to adjust. If it rolls out, it does some tasks, it takes a long time to do other tasks, well then I think we’re much more in a world where we can adjust in the way that we have in the past.
A version of this story was also published in the User Friendly newsletter. Sign up here so you don’t miss the next one!