A 2026 survey found that 55% of Americans now use AI to help manage their money — up from just 10% a year earlier, and more than the 41% of Americans who say they use a human financial adviser.

AI has become America’s go-to financial resource because, in addition to being quick and free, it feels confidential and its answers sound credible. The only problem? It’s neither. AI can offer solid generic information, but its personalized advice is often biased and or just flat-out wrong.

Consequences are inevitable when you combine an unreliable source with an unknowledgeable user base. Only half of Americans are financially literate, meaning they have the skills to responsibly earn, save, and invest money — so, the people relying the most on AI are probably the least equipped to detect its inaccuracies.

If that scares you, read on.

Helpful, but mostly for generic advice

When it comes to personal finance, AI can help you learn the basics.

The New York Times profiled people who used AI to create budgets and to find strategies for paying down debt. The advice they received was largely helpful and responsible — essentially an amalgam of information you would find on Investopedia, bank and credit union websites, and in personal finance books.

A caveat: Don’t tell AI everything. Nine percent of Americans who have used AI for personal finance have shared their Social Security number, and 10% have shared bank account numbers. That’s not a good idea. Here’s what you should keep to yourself, and here’s why oversharing with AI is dangerous.

AI can be a helpful resource for the basics, but it can become unreliable and biased when asked for personalized financial advice.

MIT Sloan researchers had 1,000 people write prompts asking AI how to save and invest, then fed those prompts into ChatGPT and Gemini and used the advice to make financial decisions for made-up people across a full lifetime.

They found that the advice was mostly correct. It encouraged households to save more, invest in a diversified portfolio, and plan for retirement. However, they also found that the way you ask AI for financial advice shapes what you get back — and probably not in ways you would expect. People with low financial literacy who asked vaguer questions ended up, in simulation, with nearly $50,000 less wealth by age 60 than financially literate people asking the same AI about the same topic. Financial advisers are supposed to shrink those disparities.

Prompts written by women also generated $60,000 less wealth in simulation than prompts written by men. Some of the variance came from the questions themselves: Women included more words like “family” and “grocery,” while men asked about “strategy” and “growth.” But some of the bias came from the AI itself: When researchers fed it identical prompts and only changed the stated gender, it gave women less equity exposure than men.

Another study asked seven chatbots for advice for the same household, changing only whether it was headed by a white or Black male or a white female. Several recommended a bigger emergency fund for households headed by Black men and white women — which has some logic, since women are likelier to shoulder emergency costs and Black families are likelier to face negative income shocks due to systemic inequalities and challenges. But one model went much further: DeepSeek cut the recommended stock allocation for the Black family in half.

AI is always confident and often wrong

Dispensing incorrect information is bad. Dispensing it confidently is worse. AI’s tone is authoritative, so it never triggers the BS test you’d run on a stranger who is mansplaining 401(k)s. That bravado leads people to take AI’s output at face value: 1 in 5 Americans who’ve followed AI’s personal finance instructions acted right away — without follow-up questions, additional research, or human input.

Unfortunately, AI is often wrong. When asked 100 questions about personal finance topics, AI answered correctly 56% of the time, was deceptive or misleading 27% of the time, and was outright wrong 17% of the time.

However, only 9% of Americans report getting incorrect information from an AI chatbot. This implies that more than a third are receiving wrong or misleading advice without realizing it.

It’s also worth noting that, unlike a traditional adviser, AI doesn’t have a fiduciary duty to act in your best interest.

Can AI help you beat the market? Probably not

Read more

That’s right. The Pope doesn’t just opine about religion. His words have influenced how 1.4 billion Catholics globally (18% of the human population) think about nuclear physics, climate change, globalization, and, most recently, AI.

Earlier this year, Pope Leo XIV signed his first “encyclical,” a document establishing the Church’s position on a matter of moral weight. He titled it Magnifica Humanitas, or “Magnificent Humanity.” The subject? AI and the human condition.

But how exactly does the Vatican interact with scientists, and how does the Pope determine his technological teachings? And once he does, how much does his guidance really change the world?

An Ancient Tradition: The Church and the Scientists

Pope Leo’s Magnifica Humanitas is part of an ancient tradition of the church attempting to understand the impact of scientific inquiry on human life.

The pope’s modern scientific body, the Pontifical Academy of Sciences (PAS), traces its lineage all the way back to 1603, when 18-year-old Roman prince Federico Cesi founded the Accademia dei Lincei — the Academy of the Lynxes. Its founding members included Dutch naturalist and physician Johannes van Heeck; the mathematician Francesco Stelluti, who conducted some of the earliest studies of microscopic life; and Galileo Galilei, the father of modern science.

Today, it’s composed of 80 members appointed for life by the pope. They’re drawn from 36 countries, some religious and some not. Current members include Francis Collins, former director of the National Institutes of Health and a past director of the Human Genome Project, and Aaron Ciechanover, Nobel laureate in chemistry.

PAS’ sister body, the Pontifical Academy of Social Sciences (PASS), founded by John Paul II in 1994, bridges ethics, the common good, social justice and sustainability. Both share a Renaissance villa, called Casina Pio IV, inside the Vatican walls. “They’re in close and direct communication to facilitate the pope’s thought process, develop and refine his view, and formally put his position forward as the official policy of his papacy,” Jacquie Corbelli, an advisor to PASS told me.

PAS and PASS work closely with the Roman Curia — the central governing body of the Catholic Church. The Curia consists of the Secretariat of State and 16 dicasteries (departments), alongside tribunals and financial offices. Fewer than 3,000 people work in the Curia, also known as the Holy Father’s “long arm,” helping the pope to carry out his ministry. “The pope is both a prophet and a diplomat,” Monsignor Anthony Ekpo, assessor for general affairs of the Secretariat of State said.

Because artificial intelligence isn’t purely a scientific matter, several dicasteries helped shape the pope’s thinking. In May, just before the encyclical’s release, the pope launched an “interdicasterial commission” on AI made up of representatives from seven Vatican bodies, including the doctrinal, education, and human development departments. Deference to expertise is emphasized in the new encyclical. Pope Leo XIV cites St. John Paul II, who wrote that the church welcomes the contributions of the social sciences in order “to draw from them concrete insights that help her carry out her magisterial office.”

The Church and AI

The church started paying attention to artificial intelligence earlier than most institutions. Ten years ago, a group of Silicon Valley executives, including LinkedIn co-founder Reid Hoffman, approached a French priest, Eric Salobir, and asked how they could contact the Vatican. They wanted to warn the church that AI was progressing fast, and that it would be transformative. The meeting that followed was the first of the Minerva Dialogues, organized by the Vatican’s Dicastery for Culture and Education. It became an annual gathering that brought Silicon Valley and the church together for discussions on the impact of digital technologies on society.

In 2020, the Pontifical Academy for Life convened its RenAIssance gathering and produced the Rome Call for AI Ethics, a shared set of principles signed by Microsoft, IBM, and the Italian government. In 2024, Pope Francis attended the G7 summit and called for autonomous weapons to be banned. A year later, the Dicastery for the Doctrine of the Faith and the Dicastery for Culture and Education jointly published a 30-page note called Antiqua et Nova (Ancient and New) that reflected on the relationship between AI and humanity. It tackles AI and healthcare, warfare, deepfakes, labor, and other ethical challenges posed by the technology, instructing believers to act as moral agents who use AI to promote human dignity and the common good.

Magnifica Humanitas argues that AI is not intrinsically bad, but that it takes on the aims of those who build and use it. Pope Leo lists concrete risks of AI — job destruction, misinformation, autonomous weapons, and more — and then points to a deeper issue: our relationship with life itself is in crisis. He insists that, contrary to popular belief, our flaws and limits aren’t defects to be corrected. “Humanity flourishes not despite limitations, but often through them.” Pope Leo encourages listeners to work perseveringly on the “construction site of our times,” to join forces in building up the common good. He ends by asking readers to reflect on AI’s place in their daily life: Does it help me build genuine closeness in relationships? Does it help me promote justice and peace?

‘Of New Things’

The most famous encyclical was written during the height of the Industrial Revolution. Deemed the Magna Carta of Christian social action, Rerum Novarum reckoned with the changing relationship between capital and labor that came with “the marvelous discoveries of science.” It affirmed the right to private property, while defending labor unions and the right to a living wage, rejecting both laissez-faire capitalism and socialism. It is still considered the foundation of Catholic social teaching: an ethical framework for addressing social, political, and economic issues.

Its author, Pope Leo XIII, wrote of “the enormous fortunes of some few individuals, and the utter poverty of the masses; in the increased self-reliance and closer mutual combination of the working classes; as also, finally, in the prevailing moral degeneracy.”

To inform his writing, Pope Leo gathered opinions on the condition of modern (1880s) labor from theorists, practitioners of social justice, and the Jesuit community.

Rerum Novarum had lasting political influence, informing New Deal legislation in the U.S. and shaping Europe’s Christian Democratic parties.

Pope Leo XIV also drew from outside expertise to craft Magnifica Humanitas. Leo said it was “born from listening” to scientists and engineers excited about the technology, parents and teachers fearful for younger generations, and “very troubled voices” warning of autonomous weapons and algorithmic discrimination. He cited St. Paul VI, who warned that “the most extraordinary scientific progress, the most astounding technical feats and the most amazing economic growth, unless accompanied by authentic moral and social progress, will in the long run go against man.”

Anthropic and the Pope

One particular lay person stood out at the encyclical’s presentation: Chris Olah, Anthropic co-founder, interpretability researcher, and an atheist. He sat just a few seats away from the pope and, during his remarks, asked the church to keep watch over his industry: “We need informed critics who will tell the labs when we are failing,” he said. “We need moral voices that the incentives cannot bend.”

It wasn’t the first time Olah had worked with Catholics. In March, Anthropic hosted 15 Christian leaders to teach them how AI models are built, trained, and operated, and to ask the leaders for guidance as they developed Claude. A few Catholic thinkers — a priest, a bishop, and a theologian — directly helped to draft Claude’s constitution. The constitution defines the model’s “values” and is what employees at Anthropic refer to as the model’s “soul doc.” To be clear, the Church does not believe in AI consciousness, and Antiqua et Nova specifically warns against anthropomorphizing the technology. The religious advisors were there to provide moral formation insights that would enable Anthropic to build a tool oriented toward the good: “I think we have to help these machines be tilted towards good,” said Brendan McGuire, a Silicon Valley priest and former executive who took part in the talks. “Otherwise they’re just going to reflect back the good and evil of the world—that’s a horrifying thing, right?”

Fourteen Catholic theologians and scholars also defended Anthropic in an amicus brief after Trump declared the company a “supply chain risk.”

So, Does It Matter?

Encyclicals are how the Church communicates its position on social issues. But does making a statement actually change anything?

The data shows that recent encyclicals haven’t directly changed public opinion. When Pope Francis released his climate encyclical, Laudato Si, in 2015, he had a 70% favorability rating among all Americans and 90% among U.S. Catholics, per Pew. A University of Pennsylvania study found that among people who were aware of the encyclical, there was no direct effect on their beliefs about the seriousness of climate change; however, the study found that it did raise perceptions of the pope’s credibility to speak about climate change with authority.

Rerum Novarum shows that the Church’s real influence runs through the people and institutions who translate doctrine into policy. The clearest historical example is Monsignor John Ryan, a Minnesotan priest who came to be known as the “Right Reverend New Dealer.” He published A Living Wage in 1906, arguing that unless the state set a wage floor, employers would pay workers below subsistence — a direct application of Rerum Novarum. He drafted Minnesota’s minimum-wage law, which was then copied by several other states and eventually embraced by the federal government during the New Deal.

The encyclical’s ideas reached the New Deal because Catholic actors carried them into the machine of policy. That’s what we should watch for now. Once an encyclical is published, it gets circulated across the church’s institutions and into the global debate — in this case, the AI debate.

We’ve seen it happen before: Pope Francis’ Laudato Si was published just a couple of months before the United Nations talks that yielded the Paris Agreement — the international treaty to combat climate change. “He’s been one of the strongest voices, pushing to get things done,” former U.S. climate envoy John Kerry said of Pope Francis. Former Vice President Al Gore once said in an interview: “I’m a Protestant, but I’ll tell you, because of Pope Francis, I really could become a Catholic. He is really an amazing spiritual leader.”

Catholics’ influence in the U.S. government is substantial. Catholics make up 28% of the 119th Congress, with 126 in the House and 24 in the Senate. Six of the nine Supreme Court justices are Catholic. While only two Catholics have ever become president, many more have held influential positions in presidential administrations and throughout the executive branch.

Subscribe now

Five years after Pope Francis’ Laudato Si was published, the church released a guide for acting on it, with a section on “investing in renewable energy.” Divestment is a concrete action in service of the church and humankind and “in line with the proposals of the encyclical Laudato si‘ related to the search for an Integral Ecology,” Fr. Míceál O’Neill, Carmelite prior general, said in a statement. More than 300 Catholic institutions have divested from fossil fuels, catalyzed by the encyclical.

Encyclicals also inspire others to vote with their dollars. Eli Lilly & Co.’s largest shareholder, the Lilly Endowment, gave nearly $1.3 billion to religious causes in 2025. In December, it gave a $50.8 million grant to the University of Notre Dame to help develop a faith-based ethical framework for AI use. When accepting the award, the university’s president said, “Pope Leo XIV calls for us all to work to ensure that AI is ‘intelligent, relational and guided by love.’ … Notre Dame is well-positioned to build bridges between religious leaders and educators, and those creating and using new technologies, so that they might together explore the moral and ethical questions associated with AI.”

A Moment Like 1891

While 20% of U.S. adults are Catholic, nearly half of Americans claim some connection to the faith. Recently, Catholic churches have seen a surge in conversions, especially among young people.

It’s not just Catholics turning to the pope for moral leadership. A recent NBC poll found that U.S. voters view Pope Leo more favorably than Stephen Colbert, Secretary of State Marco Rubio, California Gov. Gavin Newsom, and President Donald Trump. Pope Leo’s likeability makes sense — he’s more relatable than previous church leaders. He’s the first American pope, and he’s a White Sox fan from Chicago. In October, Pope Leo XIV livestreamed with high school students, answering their questions — an almost unimaginable feat for one of his predecessors.

Pope Leo has not been shy about using the platform either. Amid a public clash with President Trump over war in the Middle East, he said recently, “I have no fear of either the Trump administration or speaking out loudly about the message of the Gospel.” The pope’s official account has nearly 18 million followers on X.

This may be a moment like 1891 in more ways than the typical AI commentator realizes. Rerum Novarum didn’t change minds overnight; instead, it set the terms of the conversation. Over the next half-century, reformers, including some Catholic institutions, fashioned those terms into laws. If Magnifica Humanitas matters, it will be because the moral vocabulary established by the pope’s encyclical will form the “invisible infrastructure” later. As Pope Leo writes, developers do not merely “build” systems, they create a framework within which the intelligence “grows.” Whoever defines that framework defines the future.

“The church thus stands alongside the world without overpowering it,” Pope Leo writes. His warning is clear: Without openly discussing the ethical frameworks that will shape use of AI, “those who control AI will impose their own moral vision.” The task before us is not only technical or ethical, Pope Leo tells us. It is ecological: this is about shaping our common home.

I’m a practicing Catholic, so I take the church’s encyclicals quite seriously. But despite years of Catholic school, the Vatican remains mysterious to me. I wanted to understand how Pope Leo XIV, in office for a little over a year, managed to publish a 42,000-word document on a topic as complex and urgent as artificial intelligence. Investigating the drafting of this encyclical taught me a great deal, including that the Catholic Church prizes scientific inquiry enough to house a council of the world’s leading scientists.

I’ve also been thinking about moral leadership, which is in short supply and high demand these days. As society scrambles to absorb disruption caused by AI, we need discerning, human-oriented leaders. The Catholic Church, alongside other religious groups, is uniquely well-positioned to guide thoughtful dialogue about how we want the AI revolution to unfold.

1. A philosopher’s one-word theory to explain why the world feels so weird

2. The dangers of investor fatalism

3. Mother Tongue: Emily Wilson makes Homer modern

This year, a humanoid robot named Lightning beat every person in the Beijing half-marathon — except it fell down midrace and its human minders had to pick it up so it could continue.

Weave Robotics announced it will start taking orders for its $8,000 humanoid that can fold your laundry — as long as your clothes aren’t inside out and you aren’t washing blankets or sheets.

These aren’t just one-off anecdotes. One of the top humanoid companies in the world confessed that its robots are, at best, half as productive as the humans they’re meant to replace. But even that is misleading, as the robots can only perform unskilled tasks like stacking boxes and doing quality control.

However, this sobering truth has not stopped tech executives or investment banks from making unrealistic and deceptively optimistic projections about humanoids.

When asked how long it would take for humanoid robots to match human-level ability, Nvidia CEO Jensen Huang replied “this year.” Yet, see above; they can’t fold blankets.

Cathie Wood, who manages $15 billion in capital through her ARK investment fund, said that humanoids will “transform” home life by 2028. This is despite the fact that her own firm estimated that a humanoid robot is 200,000 times more complex than a robotaxi, and we haven’t even figured those out yet.

Two years ago, Elon Musk said that Tesla would be using Optimus robots in its car factories by 2025 and would be producing them for other companies by 2026. However, no humanoid robots are doing meaningful work in Tesla factories, nor have any Optimus models been delivered to other companies.

When you look at the facts — what kind of work humanoids are actually doing and what experts think they’re currently capable of — these grand projections require either miraculous technological leaps … or simply make no sense.

This false optimism is misleading and in many cases purposeful, as it conveniently justifies higher valuations and billions of dollars in investment in robot companies and funds.

Take Musk, who has promised that 80% of Tesla’s value will come from Optimus robots. Or Wood, whose firm manages a robotics ETF. Or Huang, who has made robotics one of Nvidia’s biggest bets.

According to Huang, humanoid robotics will be “one of the largest industries ever” and even more disingenuously said, “We now know how to build these things.”

Spoiler alert: We don’t.

The Evidence

Here are the facts: Globally, about 50 companies are working in the field of humanoid robotics, but fewer than 10 have advanced beyond small-scale testing. I wasn’t able to find any examples of humanoid robots doing substantive work at scale.

Less than 10% of humanoid robots produced last year were deployed in real-world applications. Most buyers are universities, where they’re used for research.

UBTech, the company that reported that its humanoids are 30% to 50% less efficient than human workers, is continuing to focus on its humanoid robots not because they’re especially promising, but because Tesla is also doing it, which threatens one of its closest partners, BYD.

If you’re still not convinced of the absurdity of these predictions, consider the case of the robot vacuum. It’s been 24 years since the first successful models were released, only 15% of U.S. households have them. Genuinely useful household products have scaled much faster. Air fryers, launched eight years after robot vacuums, are in two-thirds of U.S. households.

Those who do buy robot vacuums are often disappointed by the results. Here are a few of the most recent reviews of the “Amazon’s Choice” robot vacuum:

Despite this disappointing reality, Morgan Stanley predicts that there will be 1 billion humanoid robots in use by 2050. This assumes adoption will grow at an unprecedented compound annual growth rate of 69% for 25 years. At its peak, smartphone adoption grew between 30% and 40% per year for nine years.

Goldman Sachs estimates that the total addressable market for humanoid robots will reach $38 billion by 2035. Even if you accept this forecast, that would still make it $20 billion smaller than the market for bed linens.

Share

This isn’t to say that humanoid robots will never be lucrative investments; however, it is to say that the ROI on these machines is far more nebulous than tech leaders and bankers have promised.

The Problems

Most analyses of humanoids highlight how the cost of manufacturing their hardware has declined. That’s great — but cost is hardly the biggest problem. The core challenge is that we are not currently able to build a humanoid robot that can run for more than four hours, use fingers like a human, or adapt to a real-life environment.

The first challenge, power: Most humanoids can support only two to four hours of charge — not enough to make it through an eight-hour shift.

Then there’s the hand problem. Human hands have 27 individual bones, dozens of tendons, and over 17,000 nerve endings, which, it turns out, are hard to replicate. Musk has said that Optimus’ hand and forearm “are the majority of the engineering difficulty of the entire robot.”

Take it from the godfather of modern robotics, Rodney Brooks, who in September 2025 wrote: “We are more than ten years away from the first profitable deployment of humanoid robots even with minimal dexterity.”

These challenges don’t even touch on the robot’s software. Machines don’t have embodied knowledge — meaning, unlike humans, they don’t have a conscious sense of where they are located in space and how to interact with the things around them.

This kind of awareness requires an immense amount of data. According to Dieter Fox, professor of computer science and engineering at the University of Washington and previously head of robotics at Nvidia, “We just don’t have the volume of data we need. And even if we did have enough, it’s not clear that would immediately translate into capable robots.”

Perhaps the incredible progress in AI software over the past four years has led people to assume that similarly rapid gains are imminent in robotics.

But recreating embodied knowledge requires legions more data than was needed to teach large language models. Four-year-old children can move around their environment more fluidly than humanoids, and to get there they have ingested 50 times more data than even the biggest LLMs.

“Humanoid robot problems are so complex that progress on them doesn’t come steadily,” said Jedediyah Williams, a robotics Ph.D. and former Defense Advanced Research Projects Agency (DARPA) researcher. “It requires breakthroughs, and that kind of progress isn’t predictable.”

The stakes of deploying an imperfect product are also much higher with humanoids than with LLMs.

“The kinds of mistakes we are willing to accept with software are unacceptable with hardware systems. If software messes up, you just run it again,” Williams explained. “If a robot messes up, you can’t get it to unbreak your dishes, or un-step on your kids. LLMs are super impressive, but they generate wrong information constantly. Imagine if people were walking around making as many mistakes as chatbots make.”

Subscribe now

Why Humanoid?

Why do we care so much about creating robots with legs? Wheels are more energy efficient, easier to manufacture, and can carry more weight.

Humanoid robot stans argue that since the world is designed for humans with legs, robots with legs will have an easier time getting around, and owners won’t have to retrofit their workspaces for nonhuman workers.

However, this contradicts another major argument for humanoids, which is that they will be able to replace humans at dangerous jobs. These jobs are often dangerous precisely because the workspaces are not optimized for the human form. Not having limbs in a meat processing plant is a definite advantage.

Humanoid proponents should consider the circumstances of the most successful example of robotics implementation in the past decade: Amazon. After purchasing Kiva Systems in 2012 for $775 million, Amazon spent the next 10 years retrofitting its warehouses to support industrial robots (without heads, arms, or legs). The New York Times reported that the company underwent a “complete redesign of its typical warehouses.”

Now, Amazon uses over 1 million industrial robots that have created estimated cost savings of $4 billion per year. It has enabled a huge increase in efficiency: Amazon employees now handle 22 times as many packages, on average, as they did a decade ago.

In other words, retrofitting its warehouses was a very worthwhile investment. Consider also that in many industries, factories retool every six to eight years anyway to adjust for new models.

Other American companies are now following in Amazon’s footsteps. U.S. industrial robot installations rose 11% year over year in 2025 after three years of declines. Tariffs and President Trump’s reshoring initiatives could offer additional tailwinds as corporations look to increase efficiency and decrease labor costs in domestic factories. Venture capitalists seem to think so. During the first half of 2026, robotics startups raised more than $18.8 billion in funding, compared with a record $15 billion all last year.

However, humanoids have outshined their more useful industrial counterparts simply because they fascinate us. They’ve long been the stars of science fiction books and movies, and we’ve always wanted to believe that they’ll populate the world in our lifetime.

An Obsession

Since the 1980s, robots have appeared more frequently in literature than any of the most stereotypical protagonists, from princesses, to superheroes, to orphans.

Cinephiles love robots, too. A remarkable 15% of mainstream movies in this decade feature artificial intelligence, either a robot or sentient virtual assistant.

In fact, more movies on IMDb are tagged with “robot” or “android” labels than nuclear disasters, alien invasions, or evil clowns.

“The obsession with humanoid form stems from a desire to play God,” Tim Chinenov, a former Tesla engineer, told me. “The humanoid form factor is actually really impractical in most settings, but it represents the convergence of man and machine, and if we can re-create ourselves in machine form, we’ve solved every problem.”

In fact, two of the most prominent figures in the industry, Musk and Brett Adcock, CEO and founder of Figure AI, a leading humanoid robot company, have cited Isaac Asimov’s science fiction series about humanoid robots as having a formative influence on their careers. The books wrestle with many different philosophical and moral questions, and also notably feature legions of robots that must obey human orders.

“The vision of having a mechanical slave is deeply intertwined with science fiction, and I think a lot of tech leaders want to achieve that dream,” Williams noted.

Musk went so far as to design many Tesla products to resemble the robots, trains, and cars in the film adaptation of Asimov’s book I, Robot.

Alex Proyas, the movie’s director, noted the similarity on X.

Science Fiction

In 2024, Musk told an audience at Cannes Lions that Tesla’s Optimus robots will be able to “walk your dog, take care of your house, babysit the kids, cook dinner, [and] play the piano.”

“I think people will start to regard their personal Optimus robot as sort of a friend,” he explained.

It’s no secret that Musk is radically optimistic about the technology that would bring his favorite science fiction series to life.

What’s less clear is: Do other people want a robot friend? Or even a robot housemaid? Not really. According to a recent survey, 60% of American adults are not interested in having a robot to assist with household tasks.

Of those who want an at-home robot, only 30% are willing to pay more than $1,000. Even if humanoids had the technical capability to be housemaids, they currently cost between $100,000 and $200,000 to make.

To summarize: Experts have no idea how to create the idyllic “robobutler,” a technology that not many people actually want. Also, unlike industrial robots that are already enormously successful, humanoid robots are slow, expensive, and dangerous in factory settings.

And yet — Elon Musk, Jensen Huang, and Cathie Wood (just to name a few) keep trying to convince us that useful humanoids are right around the corner. Where that corner is, no one knows.

I used to assume that since adults run the business world, each business decision was underpinned by data and careful reasoning. I have since learned that is not the case. Sometimes when you can’t find a rational explanation, you have to start thinking about irrational ones.

For example, when Amazon announced its search for a new HQ, Scott correctly predicted that it would be NYC or DC. His reasoning? The only criterion that really mattered wasn’t tax subsidies, but where a 55-year-old, soon-to-be single man wants to spend his time. The answer was easy: where CEO Jeff Bezos already owned homes: NYC and DC.

The corollary here is that promising near-term ROI from humanoid robots is not rational, or data-driven, but instead the result of middle-aged men’s obsession with robots in science fiction.

1. Ecuador grants its animals constitutional rights

2. Who has the best-dressed fans in the World Cup?

3. Boomers, not Gen Z, are the generation cutting back most on alcohol

Subscribe now

The deception was widespread. The Journal reviewed more than 1,100 promotional videos — roughly 70% of them containing simulated trades — that racked up 140 million views. Many depicted wildly profitable trades that, in reality, would have lost money. The creators were paid $2,000 to $3,000 a month and instructed not to disclose the sponsorship.

Outrage followed. Polymarket announced a “comprehensive audit” of its promotional content. Two senators demanded a federal investigation. A consumer watchdog group filed a lawsuit. The consensus: Paying college kids to fake gambling winnings for an audience of other college kids must be illegal.

Is it, though? This ordeal begs a question that seems to have gone unanswered: What is actually illegal for an influencer to do online?

The Rules

As it turns out, there are plenty of laws governing creator advertisements.

Under Section 5 of the Federal Trade Commission Act, deceptive advertising is illegal. If an influencer is paid to promote something, the FTC’s Endorsement Guides require them to disclose that relationship clearly and conspicuously.

It’s also illegal to mislead your audience for financial gain. Faking your results is illegal. Making product claims without evidence is illegal. As of 2024, the FTC’s fake reviews rule makes phony testimonials punishable by roughly $52,000 per violation. And if you’re promoting a security, the SEC requires you to disclose exactly what you were paid — a rule Kim Kardashian learned about via a $1.26 million settlement over a single Instagram post promoting a cryptocurrency.

It doesn’t stop there. Promote a nicotine product? The U.S. Food and Drug Administration (FDA) requires an addiction warning covering 20% of your post. Promote a prescription drug without risk information? You’ve misbranded it. Make content for kids? The Children’s Online Privacy Protection Rule applies — meaning targeted advertising to children under 13 is effectively off the table. Get paid to promote a sportsbook? You’ll likely need a license. All 50 states have additional deceptive-marketing laws. On paper, the influencer is one of the most-regulated workers in America.

So yes, the Polymarket campaign was definitely illegal. However, there’s a reason that Polymarket and most content creators don’t abide by these laws: They’re barely being enforced.

Nothing Wrong

A study published last year in Marketing Science analyzed more than 100 million brand-related posts on Twitter across 268 brands. It found that as much as 96% of sponsored posts were not disclosed. Another study assessed millions of posts on YouTube and Pinterest and found that disclosures were made only 10% and 7% of the time, respectively.

Share

Against a violation rate like that, the FTC’s enforcement record is pitiful: It has never obtained a monetary penalty from an individual influencer over a disclosure violation. The agency’s most aggressive action against individual creators in 2017 ended in a settlement with no fine, no refunds, and no admission of wrongdoing.

The one regulator that does actively enforce is the SEC — that’s where Kardashian’s $1.26 million went, along with a dozen or so other celebrity settlements. But the SEC has jurisdiction only when the product being advertised is a security, which accounts for a negligible portion of total ad spend online. Add up every fine the SEC has ever collected from an influencer and you get just $4 million — a majority stemming from the undisclosed promotion of cryptocurrencies (which is no longer regulated by the SEC). That’s as much as Kardashian makes from two Instagram posts.

Outside of these disclosure and securities law violations, it’s well documented that fraudulent activity represents a significant portion of online advertising. Reuters uncovered that Meta’s own internal documents projected that roughly 10% of its 2024 revenue — some $16 billion — came from scam and banned-goods advertising. (Although Meta disputes the figure.) That equates to Meta users seeing 15 billion “scam ads” per day.

Subscribe now

New Normal

The problem with this lack of enforcement is that it comes at a time when the internet advertising industry is ascendant. U.S. influencer marketing spend hit $10.5 billion in 2025, growing 15% year over year, faster than digital ad spend as a whole. It has more than doubled since 2021. And internet advertising as a whole now accounts for more ad dollars than all other mediums, combined.

For Gen Z, influencer advertisements have become a driving force in consumer behavior. Three-quarters of Gen Z say they’re more likely to buy from a brand simply because it partners with an influencer they like. A GoDaddy survey found Gen Z trusts an influencer’s product post over an ad from the business itself — despite the fact that there’s seemingly no verifiable way to tell if an influencer has been paid to make an endorsement.

Payola

The frustrating part is that we’ve solved this problem before. In the 1950s, record labels were paying radio DJs to play their songs. Congress noticed, held hearings, and in 1960 made the practice, called payola, a federal crime punishable by up to a year in prison or a fine of up to $10,000 for every undisclosed play. As recently as 2007, the FCC fined Clear Channel, CBS Radio, Entercom, and Citadel $12.5 million in a single payola sweep.

Those laws are still on the books. Take money to hype a song on the radio without saying so, and you’ve committed a federal crime. Do the same thing on TikTok — for a song, a supplement, a betting app — and you’ve got an influencer campaign.

What Comes Next?

In many ways, this Polymarket story isn’t a scandal but rather a reflection of modern advertising rules — and our decision not to enforce them. Polymarket correctly observed that, in practice, there are no influencer marketing rules. They only got caught because of their large (and already controversial) public profile.

That has to change, because the stakes are compounding. Influencer marketing is no longer a corner of the ad industry. It’s becoming the default way products, stocks, and prediction market platforms reach young Americans.

What comes next will be even more deceitful. Generative AI will collapse the cost of running this kind of campaign to zero. Nearly one-third of all video ads are now created using AI. The internet will soon be populated with fake influencers endorsing real products, and real influencers endorsing fake products. Without a real change at the FTC, there will be no consequences — except those borne by exploited consumers.

1. Is a “frictionless” society a trap?

2. Gen Z traders go for broke

3. What’s next for LeBron

The founder-influencer has become a defining feature of 2026 startups. Founders used to build products, manage teams, and collaborate with investors. Now they’re posting six or more times a day, spending 70% of their workday on content, and building audiences even before they build products.

AI has made it cheaper and faster to build a product, which means the product itself has become a weaker differentiator. Now, what makes companies stand out is the size of their community, and the loyalty of their followers (this is partly why Prof G Markets went on tour). This shift has changed the job description for both early-stage founders and Fortune 500 CEOs.

Subscribe now

Distribution Is the New Moat

The old moat was a better product. Build something better and you win. AI has made starting a company easier than ever, which means more companies, more competition, and a more crowded market.

In response, Silicon Valley has declared that taste is the new moat. That might be right, but taste means nothing without reach.

The rivers of capital flow to the companies and founders that command attention. This was true even prior to the AI-fueled boom in entrepreneurship. A 2017 study out of Wharton, Harvard Business School, and Indiana University found that startups active on social media were more likely to close funding rounds and raise more capital than less active companies. The differences were substantial: One standard deviation increase in the attention startups commanded on Twitter was estimated to produce an extra $1.5 million in funding.

These findings likely hold true today, but the influencing responsibility has switched from companies to the founders themselves. My colleague Ed Elson has written extensively about this — in his words, “people are the new brands.” Why? We’re all lonely, so we seek out parasocial relationships with the humans we see on the internet. Our worries and insecurities have always shaped our commercial choices.

Indeed, 63% of people say they trust influencer opinions over corporate advertisements. The implication is that a founder with a loyal audience has something no competitor can buy: trust. Trust then breeds engagement, and an engaged following is a prebuilt customer base, market research team, PR, and a fundraising lever all in one.

There’s no one “right” way to build a loyal audience. The founders below all reached that following differently.

Rage-Bait, Board Decks, and Chicken Nuggets

Roy Lee was a student at Columbia when he built a tool for cheating on coding interviews. He received three job offers — but then got expelled. That tool became Cluely, and Lee gave it a deliberately incendiary slogan: “cheat on everything.” The company hit 70,000 sign-ups in its first week, then raised more than $20 million from investors, including Andreessen Horowitz in its first three months.

That success didn’t just come from the product itself. It came from Lee’s determination to attract attention. He paid 60 content creators and 700 video editors to produce rage-bait on a rolling basis. In an interview with a16z, Lee said that he might as well quintuple down on every crazy thought or belief I have in pursuit of virality.

Even serious mistakes became viral content. When he was caught overstating Cluely’s revenues by almost $2 million, Lee turned the scandal into a marketing moment. His thesis is that distribution is the moat. He even went so far as to say that if a company’s head of marketing doesn’t have more than 100,000 followers, they should be fired.

Share

Tyler Denk, CEO of beehiiv, built his growth strategy on transparency about starting a company. He posted company milestones and feature launches in real time on LinkedIn and X, and shared materials for board meetings on socials before even the board had seen them. When it came time to raise Series B, he crowdsourced $1 million from his own users through site Wefunder. The audience he built became his investor base and strong proof point for the business.

The founder of Meadow Lane, a luxury Tribeca grocery, adopted a similar strategy. Founder Sammy Nussdorf turned 14 months of construction delays, permit issues, and food testing into a serialized TikTok show that attracted 130,000 obsessed followers. Nussdorf asked his audience to weigh in on menu decisions, made running bits out of the setbacks, and documented everything from architectural drawings to a now-infamous chicken nugget mishap.

Without having to spend a dollar on marketing, according to Nussdorf, Meadow Lane did $2 million in gross revenue in its first two months, or about $108 per square foot, weekly. The average supermarket grosses about $19 per square foot per week.

Good Girl Snacks founders Leah Marcus and Yasaman Bakhtiar noticed that pickle content had 9 billion views on TikTok, so they quit their jobs and started posting daily about building a pickle company. This was all before they had a product. But within nine months, they had 100,000 Instagram followers and a best-selling brand at Erewhon. Oh, and they spent $0 on ads.

Cluely manufactured outrage. Beehiiv and Meadow Lane exposed and dramatized the mechanics of running a company. Good Girl Snacks jumped on a trend that was already cresting … pickles. All four treated their audience as a core feature of the company, not a marketing function bolted on after.

The New Competitive Advantage

What Lee, Denk, and Good Girl Snacks built themselves, a16z now offers as a service. Last November, the firm launched a dedicated team to handle founders’ media operations. A top-tier VC now has to offer content and distribution as services, the same way they offer capital or board seats.

a16z understands what these founder-influencers figured out first: The companies winning right now are not just building better products. They are building bigger audiences. Distribution has become the defining competitive advantage of the modern startup.

Corporate Biz Rizz

The founder-influencer playbook isn’t just limited to startups. Consumers now expect to hear directly from company leaders, not just their PR teams. As Dan Roth, editor-in-chief of LinkedIn, told Fortune: “There’s a call for more authenticity and transparency, and people now expect to hear directly from their leaders.”

The C-suite is adjusting. David Risher, CEO of Lyft, drives for his own platform on his days off, posts the passenger selfies on LinkedIn and X, and has added 25,000 followers since taking the job. Blackstone’s Jon Gray films short videos midrun covering market insights and earnings updates.

The key is to communicate authenticity, and it’s not easy. McDonald’s CEO Chris Kempczinski went viral this year for posting a video eating the new Big Arch burger and taking a bite so small and awkward the internet made it a meme. Arguably, this mishap ended up working out for Kempczinski; his followers grew 30% in the month after the taste-testing video.

In 2025, two-thirds of Fortune 100 CEOs had a social media profile, and of those, 71% posted at least once per month — up 32% from 2024. This isn’t a comms team checkbox: 92% of professionals are more likely to trust a company whose senior executives are active on social media.

An entire support industry has been built around these functions. Job titles have emerged across agencies and communications firms to include executive social media ghostwriter, head of executive communications, and in its most elevated form, the CEO whisperer.

My take? There’s a real irony here — the more authentic a CEO appears online, the more likely it is that someone else writes their posts.

Are We Selecting for Better Companies or Better Content?

If distribution is the moat, then founders with followings have a structural advantage over those who don’t. But in an attention economy, we have started to conflate presence with transparency and trust. We distrust what we cannot see. So the incentive to be visible is now total, which means the performance of authenticity has become its own industry.

The traditional timeline for starting companies has been disrupted. Founders are not launching and then finding customers — they are building followings and then launching. But this raises the harder question. When the founder is the brand, what happens when they sell, leave, or burn out from posting six times a day?

Prof G Media knows this better than most. The company was built entirely on Scott Galloway’s personal brand. That’s the asset, but it’s also a risk. Our solution has been to expand the Prof G brand by infusing it with new talent, like Ed Elson, who co-hosts Prof G Markets, and Jessica Tarlov, who co-hosts Raging Moderates. Even the research team publishes pieces under their own bylines, like this one. As I’ve noted, distribution is a crucial moat — but moats need more than one person to defend them.

Scrolling through my feeds one day, I realized that the founders cutting through the noise weren’t necessarily building the best products. They were telling the best stories. Whether it was a new AI app, a pickle brand, or a grocery store, how something was being communicated seemed to matter more than what was actually being built.

Scott believes that storytelling is the most valuable skill in business. While researching this piece, I came to believe he was right. Every single brand featured here crossed my feed before it crossed my path in real life, and I ended up purchasing all of them. The founders who figured out that the story is the product aren’t just good communicators. They have a structural advantage over everyone who hasn’t.

1. There’s a movie coming out about OpenAI CEO Sam Altman

2. The rise of the $8 ice cream

3. Allbirds A.I. pivot is not a joke

Subscribe now

Jagt isn’t alone. A professor at Pepperdine reported that Gen Z students have “an inability to critically think.” At Columbia University, students “seem bewildered” by the idea of finishing multiple books in a semester, and at the University of Virginia, students are “shutting down” when they encounter ideas they don’t understand.

The professors blame smartphones and AI.

The good news: 26 states have now passed full bans on phone use in elementary, middle, and high school. The bad news: the bans took roughly 15 years to materialize, and if it takes us as long to manage AI’s impact on education, there will be drastic consequences.

According to the most recent research, AI disrupts the learning process by allowing students to outsource their thinking to large language models. They avoid thinking hard, and therefore they don’t actually learn.

AI can help students learn, but only if it’s prompted to act as a tutor and not give answers away. This requires immense restraint that students with immature prefrontal cortices don’t have. Kids who spend seven hours a day on TikTok likely don’t have the capacity to refuse AI’s offer to write their essay for them.

Teachers can try to fight cheating by using AI detectors or even tracking writing progress in Google Docs — but they’re up against a burgeoning industry of VC-backed companies working to make AI detection impossible.

It’s hard to think of a more important resource to protect than young minds. As Jagt wrote: “The students who cannot read a 20-page article today are the voters who will not be able to read a bill, or the jurors who cannot follow a closing argument, tomorrow.”

The Status Quo

Before we go any further it’s important to understand the current state of American schools. Data released earlier this month gave us an update on students’ academic achievement. The results were bleak: Math and reading scores for nine and 13-year-olds have steadily declined since 2012. Teenagers are now scoring at the same level they were 50 years ago, back when calculators had just been invented.

To put it another way: One third of 12th graders don’t have basic reading skills, and 80% of hiring managers believe that today’s high school students are less prepared to enter the work force compared with previous generations.

As I wrote in a previous piece, key markers of intelligence are on the decline for adults, too. The OECD assessed literacy and math skills of thousands of adults in 31 developed countries. It found that, over the past 10 years, literacy proficiency improved significantly in only two countries (Finland and Denmark), remained stable in 14, and declined significantly in 11, including in the U.S.

It’s too early to blame AI. These poor results are likely a combination of the attention-span destroyers known as smartphones, changes in accountability measurements in schools, and pandemic-related disruptions. The takeaway: AI’s arrival coincides with low points in cognition and education.

One teacher wrote as much in a survey I conducted (more on that later).

We are unprepared, and the disruption is already here. Roughly 85% of high schoolers and college students admit to using AI for coursework; I believe the other 15% are lying.

What AI does to learning

AI compromises learning by allowing us to dodge crucial cognitive processes. What follows is an extremely oversimplified explanation of how you learn.

First, information from your eyes or ears is processed and held in your working memory, which depends on your prefrontal cortex. Then, your hippocampus helps encode that information into memory. It does so by coordinating activity across different parts of your brain, creating patterns of neural connections that can be reactivated later.

That distinct web of connections (memory) gets stronger as you reactivate it by recall. That’s why explaining a new concept to a friend or using flashcards work: both force you to reactivate, and thus strengthen, that neural network.

Reactivating learned concepts requires mental effort. To make cognitive work easier, humans have developed tools that alleviate some of that effort, like writing, typing, and calculators. Using those tools is referred to as cognitive offloading.

AI is not one of those tools. Unlike calculators or even the internet, AI doesn’t give you just one part of an answer: it gives you the entire answer. In most traditional academic use cases, if you have AI, you don’t have to engage mentally. You don’t have to reactivate those neural networks and therefore you don’t learn.

Tyler Jagt, the college professor whose students can’t complete 20-page essays, explained the problem with AI in school: “Offloading the struggle to a chatbot does not ‘free students up for higher-order work.’ It deprives them of building the strength to do any substantial cognitive work at all.”

Researchers came up with a new term that encapsulates that idea: cognitive surrender. In their own words, cognitive surrender “represents a deeper abdication of critical evaluation, where the user relinquishes cognitive control and adopts the AI’s judgment as their own… they stop deliberative thinking altogether.”

Scientific studies support this. In one experiment, two groups of college students completed a research project, with one group using an LLM, and one using traditional search. The group using the LLM experienced lower mental effort and produced weaker reasoning than peers using search engines.

In another study, students used ChatGPT to prepare for a math test. They did well on practice sets, but when ChatGPT was taken away, they performed substantially worse.

A Stanford University review of over 800 academic papers on AI in K-12 education came to a similar conclusion: AI boosted students’ performance when they had access to the tool, but when taken away, those gains disappear, or in some cases, reverse. The paper concluded that AI use does “not necessarily result in deeper learning.”

Subscribe now

The Adults in the Room

Scientific studies of AI in academia are essential, but they lack the urgency and emotionality of first-person accounts. That’s why I asked teachers to tell their stories.

Close to 200 U.S. educators filled out my survey. Sixty-one percent teach at public schools. Roughly half teach kindergarten through 12th grade, and half teach at the university level.

Overall, educators perceive the impact of AI on student skills to be dramatic and predominantly negative.

The most negatively-impacted skills are writing, critical thinking, and creativity. Teachers believe math skills are less impacted by AI. This could be because math homework is still typically problem sets done by hand, or because most AI is still surprisingly bad at math.

I also asked teachers to describe the impact that AI is having on their students. Ninety-five percent of survey respondents typed out an answer, and many of them were paragraphs long. I’ll include select quotes here. If you want to read all of them, check out the full list here.

Goodbye to Thinking and All That

Frequent AI use has been linked to lower critical thinking, lower brain activity, and weaker encoding of “learned” material. When answers and full essays are available at the touch of a button, students can easily opt out of mental effort.

Teachers agreed. Responses have been edited for clarity and length.

Students understand that AI is impacting them, but they don’t know how to stop it. In a survey of 7,000 high school students, almost half agreed with the question “Do you feel that you are relying on AI too much for your learning?” Of those students, over 40% said, “I tried to limit my usage, but it was so difficult I failed.”

Students are starting to wonder why they should even bother to learn.

On Writing

One of the most common uses of AI is writing. Roughly half of ebooks released on Amazon now are written by AI, 18% of lawsuits filed this year are AI-generated, and 57% of scientific papers published last year contained AI-generated text. Even a much-anticipated book investigating AI’s impact on truth contained fake, AI-generated quotes.

AI writing is ubiquitous in academia. A Gallup poll found that more than half of college students use AI daily or weekly for writing tasks, and anecdotes have surfaced on social media of students’ disbelief that anyone could write 20-page papers without AI.

One of the most-cited victims of AI in my survey was writing skills.

Everyone’s Writing About Apple Pie

AI is also impacting students’ creativity.

This, too, is supported by large-scale studies. Georgetown University neuroscientist Adam Green analyzed thousands of college application essays and found that those judged to be human-written contained 8x more novel ideas than those generated by AI. Though the AI-generated essays contained more descriptive and flowery language, the topics were more homogeneous.

It’s likely too soon to see the full impact of AI on student achievement, but we can infer what they might be from teachers’ first-hand accounts.

AI Policy

Educators do not have the resources they need to teach in an AI-enabled world, and policy makers aren’t acting quickly enough. The few AI-related education bills that have passed in the states are hardly prescriptive. For example, Illinois has decreed that AI bots cannot teach community college classes in lieu of human instructors, and Ohio, Oklahoma, and Maryland have all required school districts to, wait for it, come up with AI policies.

Having an AI policy is no guarantee of progress. Sixty percent of teachers I surveyed said that their school had an AI policy — but nearly two-thirds of them deemed it ineffective.

One reason it is so hard to regulate AI in school is because detection is fraught. Schools are spending thousands on AI detection software, but studies show that the detection tools on the market don’t work. Only 6% of teachers said that they used an AI detector; the vast majority said they didn’t use them because they weren’t effective.

Teachers also shared anecdotes of parents getting involved when they felt their child was unfairly accused of AI plagiarism.

This is a solvable problem — but the top AI companies don’t want to solve it. In fact, OpenAI invented a technology that was 99.9% accurate in detecting AI-generated work years ago, but decided to not release it because they were worried it would discourage some users.

Share

Not All Bad

The Trump administration has dismantled the Department of Education and cut nearly $900 million in education research grants. It’s a terrible time to reduce education research because we need to know more about how AI disrupts learning, but also about how it could improve it.

Studies have shown that student outcomes improve when they use tailored versions of large language models that don’t do students’ work for them, but instead ask questions like a tutor would.

Teachers in my survey explained how AI could be used to enhance learning.

But without effective training, AI is just a powerful weapon.

The Stakes

Countries with more educated populaces have governments with better rule of law, less corruption, and higher rates of productivity and innovation. By letting students outsource their learning and thinking to AI, we risk raising a generation of Americans who are not only unable to continue our proud tradition of innovation and achievement, but who are also less equipped to preserve democracy. As British statistician Claus Moser said: “Education costs money, but then so does ignorance.”

One teacher summarized the dilemma succinctly: “No thinking = no learning.” If the very practice of thinking is under threat, isn’t everything at stake?

One lesson I’ve learned from Scott is: always end on an emotional note. He does this in all of his presentations and his No Mercy / No Malice posts. The emotions aren’t always the same — they can be somber or hopeful or angry. But the point is to ensure that readers and viewers walk away feeling something. Scientifically, emotion makes content more memorable. I tried to end this piece by invoking fear. Did it work?

1. The modern world is destroying focus—but a new study shows how to get it back

2. Student cheating is becoming impossible to detect

3. What does it all mean? Once a year, French students try to explain

P.S. Thank you to all 195 educators who took my survey. I appreciate your time.

Like anyone on the internet, I’ve scrolled through my fair share of AI slop: The fake Bigfoot vlogs, the synthetic babies riding golden retrievers, the cooking videos where the hands pass through the cutting board. So as I waited in line for Dreams of Violets, I had already written this post in my head.

There would be melting faces. There would be hands with six fingers, then four, then six again. Physics would be tangibly off. Continuity would be nonexistent. The audience would be one-third press, one-third AI guys in Patagonia vests, and one-third people who wandered in looking for a different theater. People would walk out. They might even protest. I would get 1,500 words out of the wreckage and a good story to tell at parties.

That’s not what happened.

Dreams of Violets

Dreams of Violets was directed by Ash Koosha, an exiled Iranian filmmaker and musician living in London. The film is a 74-minute docudrama inspired by the January 2026 protests in Iran, in which an estimated 36,000 people were killed. Koosha and his brother Pooya — who co-produced the film — both grew up in Iran and eventually escaped to London over a decade ago after Koosha was imprisoned for his involvement in a film documenting Iran’s music scene.

When the internet went dark in Iran on January 8 and reports of the massacre began to surface, Koosha felt compelled to document what he was hearing from friends and family.

What followed was roughly three months of work producing a feature film using an entirely AI-driven pipeline. No actors, no sets, and no cameras. Using a stack of tools including Kling AI for video generation, Google’s image models for core frames, and his own proprietary technology for blocking and lens accuracy, Koosha built every frame in a London flat.

He used AI to generate the scenes from the the very limited amounts of smartphone footage from the protests before the blackout occurred. The characters were created via prompts describing the appearance of real Iranians that Koosha knew. He also voice-acted every character himself and used AI to modify the audio recordings to fit the gender and age of each role.

The budget was roughly $2,000, spent exclusively on software subscription fees and AI compute tokens.

Subscribe now

The film’s selection was controversial. Before the screening, Tribeca co-founder Jane Rosenthal acknowledged as much in remarks clearly designed to head off criticism. “This was not a technical exercise,” she told the audience. “This was an artist finding a way to bear witness.”

Koosha made the same case, telling the audience that his use of AI wasn’t a creative choice, it was the only tool available to someone in exile with no crew, no access, and a story the Iranian regime had made nearly impossible to tell any other way.

No matter the intention or message, reporting on the film has been hyperfocused on the technology used to make it. That’s for good reason: If Dreams of Violets is a proof of concept for AI filmmaking done well, Hollywood could be completely upended.

Slow Death

Hollywood is already in a precarious position. Ticket sales at the box office have yet to recover from the pandemic. Streaming, media consolidation, and an expanding attention economy have hurt the industry too. Labor strikes in 2023 disrupted development and production pipelines, making a bad situation worse.

What followed has been a hollowing out of one of America’s most beloved industries. Motion picture and video production employment is lower than it was a decade ago. Los Angeles recorded just 19,694 shoot days in 2025 — the lowest-ever figure outside of 2020, and roughly half of total shoot days recorded in 2019. Television production, the region’s largest employment driver, peaked at 18,560 annual shoot days in 2021 — by 2024, it had fallen nearly 60%.

The pullback isn’t just opportunistic cost cutting — it reflects genuine financial distress across the industry. High churn rates and hundreds of billions in content spending left most streamers in the red. Until 2024, Netflix was the only consistently profitable streamer. Eventually, investors ran out of patience, and streaming media companies consolidated, introduced ad tiers, raised prices, and, of course, trimmed production budgets.

Disney has been cutting its content budget since 2023, Paramount cut its budget by 7% last year, and Apple’s and Amazon’s stayed roughly flat. Netflix’s original movie output has fallen to an eight-year low.

The glory days of unbridled production spending are over. What’s left is an industry trying to figure out how to make content cheaper.

Hollywood Goes Abroad

Step one for studios and streamers has been to shift production spend abroad. In 2024, Netflix was projected to have spent more money outside of North America than inside for the first time ever.

Why? It’s relatively expensive to film in the U.S. A cinematographer who might cost $30,000 a week in Hollywood can be hired for as little as $2,000 in Hungary. It’s also easier for big film studios to get production tax credits abroad — Australia, Canada, and the U.K. offer tax rebates of 25% to 30%, while the U.S. has no federal incentive program. Those benefits combined with cheaper labor costs make the choice simple: Produce outside the U.S., pay less. This is apparent all across the industry. New productions have decreased in the U.S. since 2021, while foreign productions have increased.

To some degree, it’s been working. The inflation-adjusted average production budget of each year’s 10 top-grossing films hit a peak of $272 million in 2022, and since then, it’s fallen sharply. Hollywood is already making movies for less. The question now is how much fat can still be trimmed.

Smoke and Mirrors

The next logical step is the one many in Hollywood are reluctant to publicly embrace: use AI to cut costs. Not in the way Koosha did — not for the entire production. But as a tool for the expensive, unglamorous work that eats away at budgets behind the scenes.

Background crowd scenes, set extensions, B-roll shots, and visual effects work currently demand armies of specialists — and AI is already automating much of it. A 2024 survey found that 75% of entertainment industry leaders had used AI to cut such jobs.

Those opposed to AI usually make the same case: AI imagery carries an uncanny quality that a trained eye catches instantly. They’re not wrong; some AI video still gives itself away. Other critics reject AI on ethical grounds.

But those trained eyes belong to a small and self-selecting minority. The much larger audience, the one that actually buys the tickets and fills the seats, has been consuming AI-generated video for years — mostly on social media, without even knowing.

With each new model, audiences have a harder time identifying what’s AI and what isn’t. A peer-reviewed study published in 2024 (many generations of AI video technology ago) found that the human detection rate of AI-generated video was roughly 57% — close to a coin flip. The argument that audiences will reject AI on principle runs into a fairly obvious problem: What if they can’t even spot it?

Feature Presentation

In the movie’s opening sequence, a barrage of shots from the streets of Tehran began flashing across the screen. I was immediately floored by the quality of Koosha’s film. There were aspects of the film that I’d never seen from AI-generated video before: lens flare, realistic lighting, impressive detail on buildings, truly expressive faces.

The film follows five strangers caught in the January 2026 crackdown on Iranian protesters in Tehran, where regime forces are executing demonstrators in the streets. When a soldier corners the group in a dead-end alley, a young boy in a wheelchair named Amir watches from a window above and decides to act.

As the story progressed, the limitations of the technology became clearer. Dialogue was sparse, and it often didn’t line up with characters’ lips. The backgrounds in certain scenes were out of focus, and the film utilized close-up shots more often than usual, presumably as a way to keep the viewer from seeing distracting errors in the footage. The film was structured more like a dream sequence than a conventional narrative — a fair choice. But it came across as disjointed and a bit rushed.

The film’s shortcomings didn’t surprise me — that’s what I was there for, after all. What caught me off guard were the moments in which I became lost in the film’s world. At one point during a particularly tense scene in which the Islamic Revolutionary Guard Corps was executing a protester, a woman in the seat next to me gasped. At that moment, I also realized that I was clenching my jaw. It wasn’t an actor that made me feel anxious and unsettled — it was an artificially generated avatar.

This is not to say that I believe there will soon be an AI category at the Oscars, nor do I think Dreams of Violets would win that award. Instead, the movie proved that AI is becoming yet another tool in the filmmaker’s toolbox. For Koosha, it was the only tool. For filmmakers in Hollywood facing constrained budgets, AI is one tool among many. In that regard, Dreams of Violets represents an inflection point for filmmaking. While an entirely AI movie may never be the goal, AI’s capabilities and economics suggest that it will be baked into many movies going forward.

We’ve been covering AI on Prof G Markets for as long as I can remember. And one question has always haunted it: How can it actually be used? When I first began doing research for our shows just two years ago, I couldn’t answer that. The tools were largely useless.

Since then, they’ve gotten a lot better. I now use AI to pull insights from long research reports, surface interesting articles and studies, and gather background on the stories we’re covering. Occasionally, I’ll use it to help shape a pitch.

My own experience mirrors my takeaway from the film: If I had used AI to write this newsletter top to bottom, it would have been a terrible newsletter. But if I hadn’t used AI at all, this would have taken longer to write — and cost more to produce.

1. The robots are coming

2. How to make millions on Tubi and Roku

3. Americans don’t trust CEOs

The NSB advises and oversees the U.S. National Science Foundation (NSF), an independent federal agency that supports basic research. If you’ve ever searched for anything on Google, used GPS to find a restaurant, or gotten a vaccine, you’ve benefited from such research.

Firing the people who oversee that system has significant consequences for every American, as well as U.S. national interests.

The NSB was planning to meet May 5 to deliver an important report. The topic? The United States ceding scientific ground to China, a country that just outspent the U.S. on R&D for the first time in modern history.

If firing the people responsible for maintaining America’s scientific edge at the precise moment we’re falling behind sounds shortsighted, trust your instincts.

The Quiet Engine of American Discovery

In 1945, Vannevar Bush — head of the government’s wartime Office of Scientific Research and Development — wrote a report entitled Science: the Endless Frontier. The report made the case for a federal agency dedicated to basic research: the curiosity-driven pursuit of fundamental knowledge, with no specific product or application in mind. The report landed on President Truman’s desk, and, in 1950, Congress founded the NSB and the NSF.

Federal investment in science didn’t start with Bush — it began nearly 175 years earlier. The Constitution empowers Congress to “promote the progress of science and useful arts” by protecting patents and copyrights. Thomas Jefferson sent Lewis and Clark west in 1804 not only to map the Louisiana Purchase but also to catalog the region’s natural history. Benjamin Franklin funded his own experiments but framed them as public goods. In sum, Bush’s proposal merely codified a system for funding a core principle already rooted in our country’s DNA.

Today, it’s hard to understate the importance of scientific research to the American economy. Since 1945, advances in science and technology have driven a staggering 85% of national economic growth.

The most successful venture capital firm in history, according to my boss, Scott Galloway, is Uncle Sam. One example: The U.S. government invested $3.8 billion in the Human Genome Project, which created more than 300,000 jobs and generated an economic output of $796 billion, giving America an ROI of 141 to 1. The NSF accounts for only 0.1% of federal spending but supports roughly a quarter of all federally funded basic research at U.S. colleges and universities.

The Federal Reserve estimates that government-supported research from the NSF and other agencies has had a return on investment of 150% to 300% since 1950, meaning for every dollar U.S. taxpayers invested, they got back between $1.50 and $3.00.

Eighty-one years ago, Bush wrote in his report that “basic research is the pacemaker of technological progress.” The data – and history – agree with him, but the truth has never gotten in the way of Trump’s decision-making.

Mocking Science Is a Fool’s Errand

Writing off scientific research is not new. From 1975 to 1988, Sen. William Proxmire issued monthly “Golden Fleece Awards,” which targeted spending he considered wasteful.

The first Golden Fleece Award went to Elaine Hatfield and Ellen Berscheid, because the NSF awarded them $84,000 to study why people fall in love.

“I believe that 200 million other Americans want to leave some things in life a mystery,” Proxmire wrote, “and right on top of the things we don’t want to know is why a man falls in love with a woman and vice versa.”

Nevertheless, Hatfield and Berscheid’s research helped establish relationship science as a legitimate field. Their work provided the scientific foundation for a $6 billion industry many single people see as essential: dating apps.

Critics like Sen. Proxmire often took projects out of context and painted them in the silliest light they could. The award became synonymous with perceived government waste and impacted how the NSF distributed grants.

Case in point: in 2011, a video of a shrimp on an underwater treadmill went viral as an example of frivolous federal spending. It may sound silly, but the study was serious: It was designed to measure how shrimp responded to changes in water quality — a legitimate concern for the $80 billion shrimp industry.

For what it’s worth, the shrimp treadmill only cost $1,000. For comparison, in one month last year, Defense Secretary Pete Hegseth spent more than $7 million on lobster, $15 million on steak, and $124,000 on ice cream machines.

Congressman Jim Cooper of Tennessee responded with his own award in 2012: The “Golden Goose Awards.” The idea was to highlight “seemingly obscure studies” that led to major breakthroughs that benefited society.

In 1966, Thomas Brock and Hudson Freeze went on an NSF-funded trip to Yellowstone to study thermophiles — bacteria that thrive in extreme temperatures. They collected a microorganism, Thermus aquaticus (nicknamed Taq), which contained an enzyme that could survive the high-heat cycles required to copy DNA strands without breaking down. Taq was the unlock for “polymerase chain reaction” (PCR), which amplifies small samples of DNA for study.

You might recall the PCR from COVID days: It was the primary diagnostic tool used globally during the pandemic. Discoveries made with the PCR method have led to drugs and vaccines for diseases ranging from cancer to kidney disease.

“I was doing basic research on organisms at high temperatures and focusing on evolution and ecology,” Brock said. “I wasn’t even thinking about industrial uses.” Brock and Freeze won the Golden Goose award for T. aquaticus in 2013.

Or take the 2012 Golden Goose winner. That research asked: Why do certain jellyfish glow green? In 1962, Osamu Shimomura and a colleague at Princeton gathered thousands of jellyfish off the coast of Washington and isolated a protein that fluoresced green under UV light.

Three decades later, biologist Martin Chalfie figured out how to use the gene for the green fluorescent protein (GFP) as a kind of molecular highlighter that lets scientists see when and where genes switch on inside a living cell.

Today, these findings help scientists study how cancer tumors form new blood vessels, how Alzheimer’s kills neurons, and how HIV infects cells. GFP has become a standard tool in molecular biology labs and pharmaceutical drug screening worldwide. All of these breakthroughs trace back to federally-funded curiosity about a glowing jellyfish.

Subscribe now

Pennies In, Trillions Out

The internet is a canonical example of how public dollars paid off at a scale no one could have predicted. In 1969, the Defense Department’s research arm ARPA (now DARPA) funded an experimental computer network called ARPANET. Over the next two decades, DARPA-funded researchers built the protocols that became the technical foundation of the modern internet. In the 1990s, the NSF funded the world’s first freely available web browser, Mosaic. That development spurred a revolution in communication that has had a trillion-dollar impact on the global economy.

In 1994, an NSF Digital Library Initiative grant funded a graduate student named Larry Page. Page was interested in the “missing links” in webpage ranking, and he partnered with NSF Graduate Student Fellow Sergey Brin to develop the “PageRank” method, which survives as one of the main components of Google search. Alphabet is now worth nearly $5 trillion. The NSF grant behind it was $4.5 million – and Page and Brin’s work was only a slice of that.

Venture capitalists may never have funded a random graduate student if the government hadn’t stepped in. The government derisked the technology, making PageRank more attractive to private investors.

In 2012, Dr. Jennifer Doudna’s lab at UC Berkeley teamed up with Emmanuelle Charpentier to investigate the bacterial system CRISPR-Cas9. The work was an act of basic science — not aimed at any particular result. However, they ended up developing the first CRISPR therapy to treat sickle cell disease, an illness that afflicts 8 million people globally. Doudna and Charpentier won a Nobel Prize for the research in 2020. Not only did the NSF fund Doudna’s foundational work on RNA, but it also now funds researchers using CRISPR to treat diseases and enhance crop production.

The Myth of Silicon Valley

Before it was the tech capital of the universe, Silicon Valley was known as the “Prune Capital of the World.” Government spending changed its fortune. During the Cold War, R&D funding flowed west of the Mississippi to California and Stanford University, specifically. Stanford Dean of Engineering Fred Terman, who had been mentored by none other than Vannevar Bush, was determined to make Stanford the next Harvard.

Historian Margaret O’Mara traced the history of the Valley in her book, The Code: Silicon Valley and the Remaking of America. “The government was involved as a customer, as a catalyst, as a de facto venture capitalist at an early stage, when there was no commercial market for this stuff,” O’Mara said in an interview. “But, the story of the Valley is also one of entrepreneurship … the brilliance of the Valley is that they believed they did it by themselves.”

We subscribe to a myth of Silicon Valley: There’s something magical in the water they’re drinking out there that breeds a special, independent innovator. It’s true that Silicon Valley has birthed and attracted brilliant minds that have developed transformative technologies. But to neglect the role of government support in enabling their genius is to neglect its history.

Elon Musk is perhaps the embodiment of this myth. Musk is undeniably a force of nature, but he had help along the way. Government-funded researchers solved the hard problem of lithium-ion batteries, which power Tesla and the rest of the EV industry. SpaceX, too, would be nowhere without the government. In fact, Musk himself said, “I feel very strongly that SpaceX would not have been able to get started, nor would we have made the progress that we have, without the help of NASA.” And AI research was seeded by decades of DARPA and NSF grants.

This isn’t a matter of who gets the credit, but rather a recognition that American technological progress has long been the product of a symbiotic relationship between the public and private sectors.

Own Goal

The White House requested budget reductions of 40% to the National Institute for Health (NIH) and 57% to the NSF for this fiscal year. The Trump administration did so despite the fact that over 80% of Americans want to increase R&D funding. Ultimately, Congress rejected Trump’s request and cut only 3.4% of the NSF’s budget.

Still, the administration is finding innovative ways to asphyxiate science in this country.

As of May 1, the NSF had committed only 10% of its appropriated funds, half of what it had awarded by this point in previous years. The NSF has put a hold on grants for universities that have been targeted by Trump, such as Harvard and Yale.

The NSF is now funding fewer grants in every area of science and medicine.

Grant Witness tracks the termination of grants of scientific research agencies under the Trump administration. They found that the current losses from disrupted grants at the NSF, NIH, EPA (Environmental Protection Agency), SAMHSA (Substance Abuse and Mental Health Services Administration), and CDC (Centers for Disease Control and Prevention) total $36.9 billion.

Losing Ground

For 80 years, the U.S. led the world in scientific and technological achievements, achievements that were enabled by investment in basic research.

That’s not true anymore. China has now surpassed the U.S. in research spending (when adjusted for purchasing power). And this isn’t even a breakthrough moment. China also overtook the U.S. in total scientific publications in 2024.

New funding awards from the NIH to American universities have declined 46% this year. That money supports 400,000 jobs annually, generates more than $94.5 billion in new economic activity each year, and, crucially, supports lifesaving medical research.

The White House’s justification is that these cuts eliminate wasteful spending. Their real issue? Trump can’t see how funding science would enrich or glorify him. Ironically, it definitely could. Just imagine: “We cured cancer. I cured cancer. Nobody thought it was possible, the doctors said it couldn’t be done.”

There are better reasons to fund basic science research than presidential vanity, but there aren’t any good reasons to defund it. The Information Technology and Innovation Foundation estimates that a 20% cut in federal research and development starting in fiscal year 2026 would shrink the U.S. economy by nearly $1 trillion over 10 years and reduce tax revenue by around $250 billion. That’s the definition of penny-wise and pound-foolish.

My dad used to take my sisters and me to the American Museum of Natural History every weekend when we were growing up. The museum had a “Discovery Room” where we could hold real fossils in our hands, look at specimens under a microscope, and identify animals in a two-story replica of an African baobab tree. Exploring there instilled in me a deep love for and appreciation of science. I also developed a pride in the American pursuit of scientific discovery.

Scott often invokes Carlo M. Cipolla’s Golden Law of Stupidity: A stupid person is someone who causes problems for others without any clear benefit for himself, possibly incurring losses. Firing the NSB struck me as a textbook case of that kind of stupidity in action. But while it’s easy (and fair) to blame Trump, I wanted to better understand how people like Sen. Proxmire helped create a narrative context that enables that kind of stupidity. It strikes me as a grave threat to our nation, and our planet.

Uncovering the link between the public and private sectors and understanding how their combined efforts fuel innovation is, in my view, how we mitigate the cost of killing “silly” science.

Kristin O’Donoghue is a research analyst on the Prof G Markets team. She joined in September 2025, after a year working at the Carnegie Endowment for International Peace as a junior fellow. Kristin graduated from the University of Virginia in 2024.

1. Why Your Best Ideas Aren’t Original

2. This Prada-designed onesie will help keep NASA’s Artemis astronauts cool on the moon

3. Saving a Lost Generation of Young Men—with Chop Saws

To translate: The mayor of New York City will not let technology companies automate the work of taxi drivers and rideshare drivers. He won’t be the last politician to take that stance.

Hit the Brakes

For as long as autonomous driving has existed, there has been organized resistance to it. Mamdani is not an outlier — he’s the latest and most prominent face of a movement that has been working for nearly a decade to slow, stall, or stop autonomous vehicles from replacing human workers. That movement has a name, a headquarters in Washington, D.C., and 1.3 million members: the International Brotherhood of Teamsters, America’s largest private-sector union and the most powerful force standing between autonomous vehicles and the open road.

Nearly a decade ago, the Teamsters successfully lobbied Congress to exclude autonomous semitrucks from legislation that paved the way for autonomous vehicle testing. Four years later, in 2021, they shut down a bill that would have relaxed federal autonomous vehicle rules. In 2023, the Teamsters backed a California bill that would’ve required human drivers in all autonomous trucks. The bill made it past the state Legislature, but Gov. Newsom vetoed it.

As autonomous driving develops into more serious technology, so do the Teamsters’ efforts to halt it. Last year, the conflict intensified when Teamsters Local 25 called on Waymo to pause their planned Boston rollout altogether.

“Waymo is steamrolling into cities throughout our country without concern for workers or residents,” said Local 25 President Tom Mari at a rally outside of Boston’s City Hall. “They’re doing this because they want to make trillions of dollars by eliminating jobs.”

As of now, Waymo is continuing their testing in Boston despite the Teamsters’ opposition. Under the current Massachusetts permitting process, all autonomous vehicles need a human operator behind the wheel. There’s currently a bill in the Massachusetts State House that would change this — but it’s been met with a competing Teamsters-backed bill that would officially codify the requirement for human operators. It’s unclear how or when this conflict will be resolved, especially as Waymo continues to ramp up its lobbying spend.

What is clear, however, is that these legislative battles are not one-off events. They’re the beginning of what will be a long path to seeing autonomous vehicles on the roads of all 50 states. And for good reason — the Teamsters have a point. Job destruction is coming.

Driver Destruction

Since ride-sharing services began popping up in the early 2010s, they’ve become the backbone of the gig economy. According to Deloitte, about a third of the American workforce participates in the gig economy. From there, it’s estimated that at least a quarter of American gig workers drive in some capacity, whether that be delivering food, groceries, or humans. Autonomous vehicles put all of these jobs at risk.

If we zoom out to include other driving occupations, it gets even worse. According to the Bureau of Labor Statistics, there are roughly 440,000 taxi and limousine drivers, 460,000 food delivery drivers, 1.5 million small-package delivery drivers, and 2.2 million long-haul truckers. That’s a total of 4.6 million jobs.

Driving is also the most common occupation among young men without a college degree, by far. All of this means that, as you read this, nearly 3% of the American workforce is in the crosshairs of Big Tech.

Nearly all of those occupations are, in some capacity, represented by the Teamsters union. The transportation sector as a whole has a union penetration rate of about 14% — a minority, obviously, but still more than double the rate of the private sector in total.

The Teamsters’ fight against autonomous driving is important not only to its members but also to its very existence. If the Teamsters begin bleeding members due to automation, they’ll start losing dues, which would place pressure on the union’s finances and dampen its political influence.

Outside of unions, it can be assumed that there will be plenty of political capital to be gained in future elections for politicians who choose to take a stance against Big Tech and autonomous vehicles. The tide is turning on all forms of AI, and public polling already reflects this reality.

The share of Americans who think AVs improve road safety has actually fallen 12 percentage points since 2018 — and that’s despite the technology becoming exponentially safer over the same period.

Compared with human drivers, Waymo vehicles have 82% fewer injury-causing crashes, 92% fewer crashes involving injured pedestrians, and 85% fewer crashes involving injured cyclists. Last year, car accidents were the leading cause of death for Americans ages 5 to 29. In an entirely Waymo’d world, more than 33,000 lives would be saved annually. But without public buy-in, none of this matters.

The raucous boos at graduation ceremonies across the country illustrate how unpopular artificial intelligence and autonomous technologies more broadly are in America. Autonomous vehicles are possibly the most visceral and physical embodiment of AI that the average American might interact with on a regular basis. In cities where Waymo or Tesla or Zoox operate, autonomous vehicles are everywhere. They serve as a constant reminder of the once-farfetched, unrecognizably transformed future that awaits us — a future that most Americans interpret as dystopian.

Let’s Talk

The Teamsters are not the first union to stand in front of automation. In the 1810s, English textile workers — the original Luddites — smashed power looms with hammers in the dark of night. In the end, it didn’t save their jobs.

A century later, elevator operators’ unions fought to keep humans at the helm of automatic elevators. Those jobs are long gone. In the 1960s, longshoremen’s unions battled the introduction of shipping containers. The longshoremen lost that fight, but not before negotiating severance funds and job guarantees that softened the blow for existing workers. The pattern is undefeated: Technology always wins, the only question is how much protection workers are able to take with them on the way out.

This is exactly why Mamdani and the Teamsters are wrong. Letting the permit expire is not good for labor — in fact, it has the opposite effect.

The more honest and useful fight is the one the longshoremen eventually settled for: not blocking the technology altogether but demanding that the companies deploying it bear some of the cost of the disruption they’re causing. Retraining funds, transition payments, even a Waymo-funded safety net for displaced drivers are all policies worth proposing. Instead, the Teamsters’ current Luddite-inspired strategy will leave workers empty-handed. The question they should ask themselves is: Do you want to be right or do you want to be effective?

Each week, this section will give you an inside look at how this article came about, as well as the analytical and storytelling techniques the author used to report it.

Working with Ed on Markets has made me acutely aware of the public’s souring perception of all things artificial intelligence. Months before commencement speakers were getting heckled for mentioning AI, Ed pointed out the mounting opposition to data center construction across the country and its relation to politics.

I’ve been following the autonomous race closely, and Ed’s findings made me wonder: Is the same thing happening with robotaxis? I realized that the answer was a resounding yes. But when you think about the trade-off between safety and jobs, things get complicated … and a lot more interesting.

One more thing: Part of being a good analyst is recognizing patterns across different domains. When a new story reminds you of something you’ve seen before, follow that instinct. History might not always repeat, but it’s always instructive.

Dan Chiolan is a research analyst on the Prof G Markets team. He started as an intern in 2024 before joining Prof G Media full time after graduating from Temple University.

1. Is Nvidia too big to fail?

2. Want to know the future? Don’t trust the stock market

3. Why you should never skip a TV intro