It’s Labor Day, so here’s a post about how human labor is alive and well in the age of AI.

I live in San Francisco and hang out with a lot of tech people, both in the AI industry and outside of it. And one thing that almost everyone I know here believes is that AI’s main economic effect is to displace humans from their jobs. Most people don’t have concrete arguments for why this should be true; it’s just an article of faith. The conventional wisdom is pretty well summed up by the first line of this tweet:

In fact, AI companies themselves have spent years talking about how their inventions are going to render large swathes of humanity economically obsolete — an odd marketing pitch, perhaps, but one that seemed to reflect their honest expectations.

A lot of times, San Francisco tech people are out of step with the general public. This time, though, the public seems to agree. A recent Ipsos poll found that most Americans expect AI to compete with human workers more than it complements them. And Pew finds that this belief has even strengthened in recent years:

So basically, most people think AI is a job-killer. And yet somehow, this job-killer keeps stubbornly refusing to kill jobs. In the aggregate, the labor market is about as healthy as it’s ever been. The prime-age employment rate — the single best indicator of how many Americans have jobs — continues to hover near all-time highs:

Of course, there are lots of other things going on in the labor market right now besides AI. But most of those things — tariffs, the Iran war, etc. — are bad for employment. It’s not easy to identify some sort of positive shock that is canceling out the job-killing effects of AI.

Or maybe it is, if the shock is AI itself. Theoretically speaking, automation can create jobs just as easily as it can destroy them. Here are Acemoglu and Restrepo (2019), explaining the various ways that technology can affect the demand for labor:

Automation [can be bad] for labor because of a displacement effect—as capital takes over tasks previously performed by labor…

[A]utomation technology also increases productivity, and via this channel, which we call the productivity effect, it contributes to the demand for labor in non-automated tasks…

[T]he displacement effect of automation has [historically] been counterbalanced by technologies that create new tasks in which labor has a comparative advantage. Such new tasks generate not only a positive productivity effect, but also a reinstatement effect—they reinstate labor into a broader range of tasks and thus change the task content of production in favor of labor. The reinstatement effect is the polar opposite of the displacement effect and directly increases the labor share as well as labor demand. [emphasis mine]

In other words, automation can do three basic things. Yes, it can replace people and take their jobs. It can also make them more productive, which can both create jobs and destroy them.1 And, crucially, automation can create new jobs for people to do. Power looms replaced master weavers, but they created jobs for technicians and engineers to make the power looms work. The internet automated much of the work of travel agents, but created jobs for web designers. And so on.

People who think of AI as a job-killer might not have thought of the second and third of these. Or they may have thought of them, but simply assumed they’re not a big deal. Anecdotally, a lot of tech people think that AI will keep substituting for more and more tasks until A) productivity increases just increase the demand for AI, and B) there are no new tasks left for humans to do. AI detractors, meanwhile — like Daron Acemoglu himself — often simply assume that new tasks created by AI will be “bad tasks” like misinformation and cybercrime that hurt the economy instead of helping it.

But these assumptions simply might not be correct. AI might be creating lots of new tasks for humans to do. For example, software engineers are writing less and less code themselves. Instead, they’re spending more and more time telling AI to write code — that represents a productivity improvement. But they’re also trying to figure out what code to tell AI to write, making sure AI is writing the kind of code they want, integrating that code into products, and so on. Those are all new tasks. There are also a lot of software engineers working on improving AI itself, and on discovering new applications for AI. Those are new tasks as well.

This helps explain why in the age of Codex and Claude Code, software developer jobs have been increasing as a percentage of total employment:

Anecdotally, organizations that thought they could replace lots of their software engineers with AI ended up having to hire many of them back — sometimes at a premium.

In fact, this is a story we see throughout the economy. Alex Tabarrok recently reported on a Census Bureau survey about AI that’s been running since 2023. The Census Bureau calls companies up and asks them A) how AI affected their total employment, and B) how AI affects the tasks that workers do.

Most companies reported no change in overall employment, which could just be due to inertia. But of companies that did report a change, more reported an increase than a decrease!

And here’s the breakdown by sector:

The story was similar for tasks. Tabarrok writes:

Among firms using AI, 44% say it supplemented or enhanced work an employee already does. Ten percent say it performed a task an employee used to do. Eleven percent say it introduced a task no one had been doing.

Here’s the chart:

Rigorous research, meanwhile, sometimes finds negative effects of AI on labor demand at the industry level, and sometimes not. But at the company level, the evidence is clearer — Kharazian, Simon, and Stevens (2026) find that when companies adopt more AI, they tend to hire humans rather than replacing them. Here’s a blog writeup of their findings:

Ramp Economics Lab

We can finally say AI isn’t killing jobs

Dear Colleagues: The most important economic question of this decade asks how AI will affect jobs. Everyone wants to write that paper. Until now, no one has had the right dataset, so existing research has relied on a combination of guesses, surveys, AI exposure scores, and self-interested punditry. In fact, a recent paper from Stanford said the ideal da…

Read more

2 months ago · 63 likes · 14 comments · Ara Kharazian

And here’s a chart:

Interestingly, they find the same for entry-level jobs — the jobs that people usually identify as being most under threat from AI.

So despite Acemoglu’s skepticism, it looks like for now, the new tasks being created by AI are probably matching or even slightly exceeding the tasks replaced by AI. Of course this measure is “number of companies” rather than “number of jobs”, but the pattern is pretty clear.

The Economist, meanwhile, has a report on how AI is creating jobs, both through the “new tasks” channel and by boosting demand in areas that AI can’t yet touch — physical jobs like construction and HVAC installation. Here’s what they write about the productivity/demand effect:

The Economist estimates that AI has so far created around 1m new jobs in America. That easily exceeds the roughly 200,000 lay-offs attributed to AI since mid-2023, and appears more than enough to offset weaker hiring in many back-office roles. America’s AI infrastructure splurge has created many of them…The Economist tracked five industries at the heart of the data-centre build-out, from electrical contracting to equipment manufacturing. Since 2023 employment in them has risen by roughly 320,000 more than broader…trends would suggest…LinkedIn, a social network for strivers, estimates that nearly half a million data-centre jobs were created between 2023 and 2025 in America, with data-centre technicians and engineers among the most common recent hires…

The scramble for workers is showing up in pay cheques, too. Indeed finds that installation and maintenance jobs at data centres advertise wages about 40% higher than comparable work elsewhere…In the year to June, average hourly earnings rose more than 13% in electrical-equipment manufacturing and nearly 8% among electrical contractors. [emphasis mine]

And here’s what they write about new tasks:

AI is also creating a new class of white-collar jobs. Engineers build the models, data annotators label their inputs and judge their answers, “forward-deployed” engineers adapt them for customers, and newly minted “heads of AI” decide what companies should do with the technology. Some of these roles barely existed until recently. Many are quickly growing in number. Postings for heads of AI, AI engineers and directors of AI have roughly doubled since 2023-24, according to LinkedIn…

Preliminary research by Gad Levanon, chief economist at the Burning Glass Institute…reckons roughly 1% of professional jobs are now “AI jobs”…[P]rofessional occupations closest to the AI boom—engineers, software developers, mathematicians and data scientists…have added roughly 730,000 jobs above trend in recent years[.] [emphasis mine]

What about specific occupations? Technology has certainly destroyed many specific types of jobs over the centuries — there are (basically) no more elevator operators, human telephone operators, or people who do manual typesetting for printing.

And yet in recent decades, we haven’t seen as much of this sort of occupational destruction. For example, a lot of people thought the internet would kill travel agents. And while the industry was hit hard, there are still plenty of travel agents left:

The reason is probably that the job of “travel agent” is much more flexible and “messy” than older types of jobs like elevator operator; travel agents do a whole lot of different tasks, so they’re harder to replace than people who just stand there and press a button. That makes modern jobs harder to replace entirely.

It’s a good bet that AI will eventually make some occupations obsolete. But so far, despite awe-inspiring progress in model capabilities, it’s extremely hard to find occupations that have seen significant replacement by AI. Top AI researchers who famously predicted the end of human radiologists saw their predictions get confounded. Truckers, too, are doing just fine.

The most impressive example might be translators. It seems pretty obvious how AI could replace human translators, and yet it hasn’t done so yet:

Here’s a chart:

If you could go back to 2022, and tell people that in four years, AI would be solving frontier math problems, but we’d still have the same number of people working as translators, how many would have believed you?

It turns out that it’s very natural for people to overestimate the degree to which AI will take their jobs. Hartley et al. (2026) have a really excellent paper called “Job Loss Fears in the First Years of Generative Artificial Intelligence”. Here’s a thread explaining the paper’s findings.

Basically, the authors find that fear of AI job replacement is extremely common:

And they find that the more people’s jobs are exposed to AI, the more they think their jobs are about to be replaced:

In fact, the more of their day people spend using AI at work, the more they’re afraid of being replaced!

And yet when the authors looked for a correlation between AI exposure and actual job loss, they found…absolutely nothing. People’s fears simply haven’t come true yet.

What’s going on? The authors hypothesize that people who use AI more start to understand its ability to replace the tasks they do at work. But as we keep finding, replacing tasks isn’t the same as replacing jobs. People keep finding new things to do in their roles at work — sometimes things AI can’t do yet, but often things that couldn’t even be done until AI made them possible!

It seems like we’re uncovering a consistent human blind spot here: People don’t actually know how they produce value at their jobs. Modern jobs are much more than a simple collection of tasks — they are pieces of a complex machine that produces value in ways that an individual worker often doesn’t even see.2 So when AI comes along and starts replacing people at various tasks, it just ends up making them more valuable as pieces of their corporate machines.

How long that situation will persist, of course, is an open question. AI leaders are starting to realize that it might take a very long time for the full effect of their inventions to be felt:

This is why the AI companies’ recent messaging pivot — many now say that AI will create jobs rather than destroying them — may be honest, rather than a cynical marketing ploy to calm public outrage.

But then there’s the question: Can this situation persist indefinitely? No one knows, of course. But my bet is that while many occupations will eventually be mostly replaced by AI, humans will still have plenty to do. I’ve argued that in order for AI to start replacing human jobs wholesale, it’ll have to get much more agentic — which will make it inherently more unreliable from a human point of view. So I predict that humans will always have jobs keeping AI agents on track.

Even if I’m wrong, though — even if the AI job apocalypse does eventually come — it doesn’t seem like it’s coming soon, and it certainly isn’t here right now. Everyone keeps thinking that AI is a job killer, and AI keeps on refusing to be what everyone expects.

Happy Labor Day!

Subscribe now

Share

When Paul St. Louis voted as a juror in 2017 to convict Frederick Turner for his role in a methamphetamine dealing ring, St. Louis had no idea that prosecutors would stack the charges in such a way that Turner, a first-time offender, received a 40-year mandatory minimum sentence.

The judge presiding over the trial declared Turner's sentence "excessive" and "wrong," but had no choice but to impose it. Less than a year later, Turner, 38, was found dead in his cell at a high-security federal prison in Colorado.

Ever since, St. Louis has been haunted by Turner's death, and the knowledge in hindsight that there may have been more he could have done in his role as a juror.

"If I could go back in time, and if I knew Frederick faced 40 years, I would vote to acquit him," St. Louis said in a new jury-education campaign launched this week by the Cato Institute, a libertarian think tank. "I could have asked what sentence would be imposed. I could have drawn my own conclusions if no one would answer. And I could have voted my conscience."

The nonpartisan initiative, Your Verdict Counts, aims to educate jurors on three things the Cato Institute says they should know but are rarely, if ever, told: that jurors have the right to decide if a law is unjust or being applied unjustly, the right to ask what the consequences of a conviction will be, and the right to vote their conscience based on those factors without fear of punishment.

"Jurors are not just called to go in and check boxes and determine if somebody technically violated the law," says Matthew Cavedon, director of the Cato Institute's Project on Criminal Justice. "Their job is to use their conscience and their sense of justice, to figure out whether somebody deserves to have their lives, their liberty, or their property denied. That's a completely different task than the one that judges and lawyers often insist is all that a jury is supposed to do."

In addition to the website, the Cato campaign will include a series of in-person events in Richmond, Virginia. Cavedon said the Cato Institute chose Richmond for the initiative because Virginia allows convicted defendants to ask to be sentenced by a jury, rather than a judge—a rare if not wholly unique feature of its criminal justice system. Richmond also has a particularly zealous defense bar.

The act of acquitting or refusing to indict a criminal defendant based on moral objections to the laws or charges in question, regardless of whether the defendant is technically guilty, is commonly called "jury nullification," and it's a controversial subject in the legal profession.

As George Mason University law professor Ilya Somin wrote at The Volokh Conspiracy in 2018, nullification undermines the rule of law in a system where the criminal codes are more or less uniformly applied, but in the real world it has become, unfortunately,  "a counterweight to the enormous discretionary power already wielded by government officials."

The Cato campaign conspicuously avoids the term nullification, which Cavedon says gives a rebellious shade to a proper and traditional check on government overreach.

It's more important than ever for juries to exercise that power, the Cato Institute argues, because juries so rarely can anymore. A Pew Research Center report found that in 2018, only 2 percent of federal criminal cases went to trial. Lengthy mandatory minimum sentences give prosecutors so much leverage over defendants that turning down a plea deal and going to trial becomes an irrationally risky choice. Defendants who turn down plea deals and get convicted, such as Turner, get hammered at sentencing, a phenomenon that criminal justice advocates call the "trial penalty." The result is that the constitutional right to trial has largely vanished.

The Cato Institute argues that emboldened juries would be a safeguard in a justice system that has mostly taken away judges' ability to consider individual defendants at sentencing.

However, although jurors can't be punished for voting their conscience, it's practically verboten to talk about inside courthouses. Attorneys are usually barred from talking about nullification with juries, and doing so can be grounds for a mistrial.

In 2019, a divided 2nd Circuit Court of Appeals panel chastised a federal judge who complained about an overzealous prosecution and said the case before him "call[ed] for jury nullification."

Over the years, activists have even been arrested and convicted on jury tampering charges for handing out flyers about nullification outside courthouses, leading to First Amendment lawsuits. In 2022, the 2nd Circuit ruled that a New York man's free speech rights were violated when he was arrested outside a Bronx courthouse for doing so.

Despite courts' deep uneasiness with the concept, the Cato Institute argues that there is a long and strong tradition of American juries using their power to resist unjust laws and charges, from the acquittal of Peter Zenger in a famous 1735 libel case to northern juries' refusal to convict defendants of aiding fleeing slaves under the Fugitive Slave Act.

Likewise, Reason's Damon Root wrote in 2018 on the originalist case for jury nullification, finding that there's "good evidence that many 18th and 19th century Americans understood juries to possess the lawful power to reject guilty verdicts when the jurors believed the underlying law was unconstitutional."

The concerns that motivated those Americans, Cavedon says, are much the same as today's.

"There's such a feeling in America today that government is distant, and that especially lawyers and judges are just part of the elite class, making decisions without the good sense of the people. That's been a concern all throughout American history, and the jury is supposed to be the antidote," Cavedon says. "The jury is literally government by the people, and we have lost something in shifting the most important function of the government, criminal punishment, to behind closed doors, to negotiations between lawyers, to a quick sign-off by a judge. That is not the freedom and the democratic order that we were promised."

The post The Right to Trial is Vanishing. Could Bolder Juries Save It? appeared first on Reason Magazine.

This month, there will be over 400 Chinese officials in New Delhi—more than double the previous delegation in 2019—to discuss the future of Sino–Indian relations. For China, those relations have never been more favorable; India’s position on the world stage has declined, though not in ways caused by China, making the prospect of resumed and increased trade between the two nations increasingly likely.

A key development came in March 2026, when the Indian Union Cabinet amended Press Note 3, a rule in place since 2020 that required government clearance for investment from any country sharing a land border with India (the largest of course being China, which shares a 3,500-kilometer border with India). The removal of this barrier proved its effectiveness by the end of August: investors had brought ₹4,896 crore ($512 million) across 29 projects. This is a proof of concept that cannot be easily ignored.

But the greater significance is in where investment did not come from: China. After relaxing a rule that primarily affected China, investment instead came from Mauritius, Korea, Japan, Singapore, and the United States.

From the two factors—the small inflow, and the source nations—a reasonable conclusion can be drawn: global funds carrying incidental Chinese shareholders had been caught by mistake, and they were now unblocked.

The conclusion that should be drawn is that formal (“hard”) blockages to trade are only ever one part of a greater picture and should be understood alongside the informal (“soft”) blockages. India’s trade problem with China is habitually described as a deficit to be narrowed by negotiation—the lifting of tariffs, increasing market access, joint ventures, and so on. It is better understood as an asymmetry that negotiation cannot alter: India’s constant attempts to address this gap seem to have made it wider.

Bilateral goods trade reached $151.1 billion in the fiscal year ending March 2026, and, of that, imports to India from China were $131.63 billion, the highest India recorded from any country; meanwhile, exports to China were only $19.47 billion. The new fiscal year is following the same trend: June’s deficit alone was $15.3 billion.

What India buys from China falls into roughly four broad categories—electronics, machinery, organic chemicals and plastics—as well as the pharmaceutical agents upon which India’s generic drug industry depends. What India sells in return is very much the raw end of things: ores and slag, mineral fuels, and marine products. Exports have been growing—up 37% in the previous year—but that growth seems less impressive when the actual numbers are considered: from $14 billion to $19 billion.

The Sino–Indian trade profile is not, therefore, one between two comparable advanced economies, but between a supplier and a manufacturer, and because the Indian economy supplies the very items that China exports to India, it becomes a self-reinforcing cycle. As Indian industry grows, and the country becomes wealthier, it will buy more of the products China produces using the very raw materials it has bought from India.

This yawning deficit is, in large part, a product of two of India’s own policies that have pulled in opposing directions.

The first is security-focused: Press Note 3, mentioned above, was first introduced at the very beginning of the COVID-19 pandemic, and was originally aimed at preventing opportunistic pandemic-era acquisitions. As a consequence, hundreds of Chinese investment applications were banned, and business visas for Chinese engineers were subject to layer upon layer of vetting, at the ministerial level. The cost of doing business simply skyrocketed. In a strict sense, it achieved its goal: Chinese direct investment collapsed to a mere $67.34 million between 2021 and 2024.

The second was an industrial policy designed to build a modern industrial base for India at speed. The first industries to benefit from this were the assembly-based industries: phones, appliances, solar panels, and so on. Naturally, these industries are resource-intensive and require more components as they grow; as a result, the trade deficit with China, the main supplier for these components, ballooned from $44 billion in 2021 to nearly $85 billion in 2024 and even further to $112 billion in 2025.

Delhi has been laser-focused on course correction since late 2025. Former cabinet secretary Rajiv Gauba proposed dismantling the visa-vetting system, and shortly after in December the government launched a digital sponsorship platform, simplifying the process and no longer requiring ministerial recommendations. On the FDI side, Press Note 2, which was approved in March and has been in effect since May 2026, automatically processed noncontrolling stakes of up to 10%, while introducing a 60-day approval window for priority manufacturing sectors (including electronics, capital goods, and solar cells).

These are real liberalization efforts, noted approvingly by the Carnegie Endowment, which wrote, “The practical argument for reform was genuine. The system was deterring not just Chinese capital but also legitimate global investment.”

The problem persists, however: the barriers to Chinese investment have been lowered, but Chinese investment has not automatically flowed through.

The forthcoming BRICS summit, held in Delhi from September 12–13, will be a real opportunity for India to address this. Xi Jinping himself will be coming, his first visit to India since 2019, with the aforementioned 400 expected delegates. The size of the delegation speaks to the seriousness of the summit and the potential for both sides (in fact, all BRICS nations) to take advantage of the global trend among manufacturers to pursue a “China Plus One” policy of diversification in response to a fragile global economy.

As it stands, trade will grow between the two nations, because both economies are growing and complement one another. Unfortunately for India, the trade deficit will grow, too, unless some serious rebalancing is made.

“Dance won’t pay well. You should choose a real job.”

“There aren’t many jobs in the dance world, and it’s really hard to get into. You should have a backup plan.”

And my personal favorite: after I shared with my high school counselor that I planned to dance as long as I could and then go to law school, she jokingly replied, “That’s just what the world needs—more dancing lawyers.”

This was the well-meaning advice I got from most of the adults in my life during my final years of high school. I was recently reflecting on that advice, now with the hindsight of more than 20 years of making a living doing things I love, and I realized where it fell short for me.

Now, who is to say whether my teenage brain would have listened or heeded their advice anyway? But as an adult, I look back at that dreamy-eyed ballerina hopeful and want to tell her this instead:

“There are many ways to make an income that will allow dance to be a large part of your life. Getting a ‘job’ in a dance company is just one of them. Learn how to teach, choreograph, judge competitions, run your own business, sell your curriculum, invest in real estate, and make dance films.”

She would probably stare at me blankly, but I would also tell her about podcasts, freelance writing, short-form video, online courses, and all the other ways people now turn skills and ideas into income. And perhaps most importantly, you don’t have to do them or learn them all right now. Build a little at a time.

You see, my high school counselors couldn’t really predict my career path 20 years in advance, and not simply because I was interested in a less traditional industry. Much of the work I eventually got paid to do didn’t even exist in the early 2000s.

I’m not alone. There is a term for the kind of eclectic working life I’ve stumbled into: a portfolio career. Although it was coined in the early ’90s, the term was mostly used for job transition for older professionals, until now. In recent years, there has been a recent increase in the use of this term as well as an increase in the number of young people following this path to build livelihoods. In fact, Gen Z is projected to work 18 jobs across 6 careers in their lifetime. In a portfolio career, rather than relying on one job, one employer, or even one profession, a person builds a livelihood from multiple skills, roles, businesses, clients, investments, and income streams.

In my own experience, very few of the things I’ve been paid to do have been traditional W-2 employment with a salary and benefits. And yet, I’ve been able to pay the bills, save, invest, and, most importantly to that teenage ballerina, do a lot of dancing.

Nobody suggested any of this to me in high school. The general consensus seemed to be that passions that couldn’t provide well-paying employment needed to remain hobbies, and that the safest path to financial security was a college degree followed by a good W-2 job.

I realize now that when educators told me dance wasn’t a good career, what they really meant was that dance wasn’t good employment, and those are not the same thing. If someone had explained that distinction to me, they might have encouraged me to learn about taxes and tax strategy, bookkeeping, business licenses and business structures, multiple income streams, investing, marketing, sales, and how to develop other skills that could also produce income.

They might have taught me how to build a livelihood instead of simply helping me choose a job. And that is something I think about a lot now as both an educator and a parent. Because if we couldn’t accurately predict the jobs that would exist 20 years from now when I was in high school, we have even less reason to believe that we can do it for today’s kids. 

AI is changing the landscape even faster, so I’m increasingly convinced that career preparation should be about helping kids construct a livelihood rather than merely selecting a career path. That requires a different set of skills, so here are some things I plan to focus on.

1. Learn to create value.

Every business is ultimately selling a solution to a problem or helping someone get a job done. When we buy something, we are essentially outsourcing a job to that thing or person. People who can recognize problems and create useful solutions will always have opportunities. You might solve a problem for an employer, a single client, or a whole group of customers, but learn how to offer your skills to others.

2. Be adaptable, and develop a variety of skills.

We need to teach kids that a career may be selected but a livelihood is constructed. This means your work might change over the course of your life, your skills may be valuable in a variety of settings, and your income streams may (and probably should) come from a variety of sources. Rather than finding “the thing” you want to do forever, become someone who can learn, adapt, notice opportunities, and figure out different ways to apply your skills and knowledge.

3. Become financially literate.

We need to question the notion that bi-weekly large paychecks equate to financial security. Without some understanding of taxes, investing, retirement accounts, interest, debt, loans, business structures, insurance, and cash flow, kids can’t understand that a large salary doesn’t always lead to wealth. We need to help them see the trade-offs between a $100,000 job and a $100,000 business, or a $60,000 job without debt versus a $100,000 job with high debt and 60-hour workweeks. Check out some of the personal finance curriculum resources offered here at the Foundation for Economic Education.

4. Develop agency.

I want kids to grow up believing that when the path in front of them doesn’t exist, they can build one. If there isn’t a job for the thing you’re good at, can you create one? If the business you want doesn’t exist, can you start it? If you can’t find a customer, can you learn how to reach one? If technology changes your industry, can you learn something new?

This isn’t to argue that everyone needs to be an entrepreneur, either. But I would like kids to understand that traditional employment is only one way or one part of constructing a livelihood. They are the CEO of their skillset, and I want to empower them to use a variety of opportunities to build the life they want to live.

The post Don’t blame austerity or Brexit for Britain’s economic woes appeared first on spiked.

ANN ARBOR, MI — A highly controversial ending at Michigan Stadium has unfortunately put a slight black eye on the NCAA's otherwise sterling reputation.