Featured

Anthropic Stock Is Falling, Wall Street Is Panicking, and Apparently Artificial Intelligence Can't Fix Human Stupidity

By a Market Observer Who Has Seen This Circus Before

Welcome to Wall Street, Where Yesterday's Genius Is Today's Idiot

I woke up this morning, November 20, 2026, to discover that Wall Street was once again behaving like a nervous Chihuahua trapped inside a fireworks factory. Anthropic, the artificial intelligence company responsible for Claude, was falling in premarket trading just one day after making its highly anticipated public debut. Technology stocks were getting hammered, Nasdaq futures were flashing red, and financial commentators were searching desperately for a respectable explanation for the industry's latest collective nervous breakdown.

Apparently, artificial intelligence can write software, analyze mountains of data, solve complicated mathematical problems, and transform entire industries, but it still hasn't figured out how to keep investors from behaving like emotionally unstable teenagers who just discovered their crush is dating someone else.

And the irony is absolutely delicious.

Here we have an industry built around the promise of superior intelligence, being traded by people who seem incapable of remembering what they believed yesterday. On Thursday, Anthropic was the shiny new object that was supposed to restore confidence in the AI revolution. By Friday morning, investors were treating the stock like it had personally offended their ancestors.

The company's shares surged more than 17% on their first trading day, closing at $97.17 after being offered at $83. That sounds impressive, and in isolation it certainly is. Unfortunately, isolation is a luxury Wall Street doesn't offer. By Friday morning, Anthropic was down approximately 6.8% in premarket trading, surrendering a portion of those gains before the coffee had finished brewing.

And naturally, everyone needed an explanation.

Was artificial intelligence suddenly worthless? Had Claude forgotten how to answer questions? Had the world's corporations collectively decided that automating expensive tasks was a terrible idea?

Of course not.

What changed was the price investors were willing to pay for the future. And that, ladies and gentlemen, is where the circus really begins.

The Anthropic IPO: A Spectacular Party With a Financial Hangover

I have always found initial public offerings fascinating, primarily because they expose the peculiar relationship between enthusiasm and rational thought.

An IPO is essentially an invitation to purchase ownership in a company whose future everybody insists is extraordinary, often at a price that assumes an impressive portion of that extraordinary future has already happened. Investment bankers dress the occasion in expensive language, executives ring ceremonial bells, analysts manufacture increasingly elaborate justifications, and ordinary investors are invited to participate in what can resemble a financial wedding reception where nobody bothers discussing the cost of the marriage.

Anthropic's debut was a major event for the artificial intelligence industry. The company had become one of the most prominent developers of advanced AI systems, competing for enterprise customers, developers, and institutional attention in a field increasingly viewed as essential to the future of computing.

Its public listing was supposed to serve as a referendum on the industry's prospects.

And initially, the crowd responded enthusiastically.

The shares opened at $111.50, climbed as high as $115.10, and eventually closed at $97.17. That is a remarkable range for a single trading session. It also demonstrates something financial television frequently neglects to mention: a stock can enjoy a successful IPO while delivering a rather unpleasant experience to somebody who bought near the opening.

Imagine being the enthusiastic investor who purchased shares at $111.50 because everyone on television was celebrating. By the closing bell, those shares were worth $97.17.

Congratulations. You participated in a historic event and received an immediate education in the difference between a company's success and your investment results.

The company raised substantial capital. The underwriting machinery did its job. Existing stakeholders gained a public market for their ownership interests. Yet the investor who purchased at an inflated opening price may have spent the evening wondering whether the revolution came with a refund policy.

This is why I refuse to confuse an exciting business with an automatically attractive stock.

They are not the same thing.

A business can change civilization while its shareholders discover that paying too much for greatness is an excellent way to lose money.

Artificial Intelligence Is Revolutionary. That Doesn't Mean Every Valuation Is Sane.

I want to make something perfectly clear before the AI enthusiasts begin sharpening their digital pitchforks.

I believe artificial intelligence is transformative.

I believe it will alter the economics of software development, customer service, research, education, professional services, industrial automation, and a variety of other industries. I believe companies that successfully deploy the technology will have opportunities to improve productivity in ways we are only beginning to understand.

What I do not believe is that recognizing a technological revolution requires abandoning elementary arithmetic.

Somewhere along the way, investors began treating artificial intelligence less like a business opportunity and more like a religious experience. Mention AI in an investor presentation and suddenly the ordinary questions about cash flow, profitability, competition, and capital requirements become annoying interruptions.

Apparently, asking how much money a company earns is terribly old-fashioned when the company promises to reinvent the human condition.

Call me primitive, but I still like businesses that eventually produce more cash than they consume.

The trouble with revolutionary industries is that the revolution and the investment returns do not necessarily arrive together. Sometimes the revolution succeeds spectacularly while the original investors endure years of disappointment.

History is full of technologies that changed the world without making every participating company a financial winner.

Railroads transformed transportation. Automobiles reorganized cities. Airlines connected continents. The internet altered commerce, communication, and information itself.

None of those achievements meant every company involved deserved an unlimited valuation.

And artificial intelligence is not exempt from economic gravity simply because its software can explain quantum physics.

The central question isn't whether AI matters. It obviously does. The question is whether individual companies can turn technological capabilities into durable profits sufficient to justify their market values.

That requires revenue, margins, disciplined spending, competitive advantages, and returns on invested capital.

These are not glamorous concepts. They don't generate thrilling headlines or inspire people to declare that civilization has reached another evolutionary threshold.

They are, however, the things that ultimately determine whether a business makes money.

And I remain stubbornly interested in that detail.

Wall Street Has Discovered That Building the Future Costs Money

One of the more entertaining developments in the artificial intelligence boom has been Wall Street's gradual realization that enormous computing infrastructure is not free.

I realize this may come as a devastating revelation.

Apparently, the servers must be purchased. The chips must be manufactured. The data centers must be constructed. Electricity must be generated. Networks must be maintained. Engineers must be paid.

Who could possibly have anticipated such outrageous expenses?

Investors spent years celebrating every announcement of additional AI investment. A technology company could declare that it intended to spend another few billion dollars on infrastructure, and the market would applaud as though management had discovered an alternative to paying taxes.

The spending was evidence of ambition. The spending was evidence of demand. The spending was proof that the company understood the future.

Until, naturally, investors began asking when the spending would produce acceptable financial returns.

That is when enthusiasm started developing a rather inconvenient relationship with reality.

The concern spreading through technology stocks isn't necessarily that artificial intelligence has stopped progressing. It is that the economic returns from the infrastructure boom may arrive more slowly, or prove smaller, than aggressive valuations had assumed.

Companies can spend billions developing computational capacity and still face declining prices for the services that capacity provides. Competitors can develop better models. Customers can demand cheaper access. Businesses can discover that implementing AI involves expensive organizational changes rather than simply purchasing a subscription and firing half the accounting department.

Technology adoption is rarely as straightforward as the promotional presentations suggest.

I can buy an exercise machine tomorrow. That doesn't mean I'll have abdominal muscles by Tuesday.

Similarly, a corporation can spend millions acquiring AI capabilities without immediately generating millions in additional profit.

The equipment is not the achievement.

The economic improvement is.

That distinction has become increasingly important as investors reconsider the extraordinary amounts being committed to artificial intelligence infrastructure.

Nvidia, AMD, and the Rest of the Technological Food Chain

Anthropic's premarket decline wasn't occurring in a vacuum. Nvidia, Advanced Micro Devices, Broadcom, and other semiconductor companies were also under pressure. Cloud infrastructure businesses and major technology companies were experiencing selling as investors reconsidered the economics of the broader AI investment cycle.

This is important because artificial intelligence has developed an enormous financial ecosystem.

Anthropic develops advanced AI models. Semiconductor companies provide computational hardware. Cloud providers deliver the infrastructure. Data center operators construct and maintain facilities. Electricity providers support increasingly demanding operations. Networking businesses connect the equipment, while memory manufacturers supply essential components.

Everyone has a role in the grand production.

And everyone would very much appreciate receiving an enormous check.

The difficulty begins when investors assume that every participant can simultaneously enjoy extraordinary growth, expanding margins, and permanent competitive advantages.

That is an ambitious financial fantasy.

Some companies will dominate particular segments. Others will compete away their advantages. Some will benefit from falling computing costs, while others will discover that falling prices are considerably less enjoyable when they are the ones selling the equipment.

I find the semiconductor industry particularly interesting because it reminds me of an old lesson from gold rushes: selling equipment to prospectors can be more profitable than digging for gold.

But that strategy has limitations, too.

If miners become overextended, financing dries up, or expected profits fail to materialize, demand for expensive equipment can weaken. The suppliers may have excellent products and substantial technological advantages, yet their growth expectations can still prove unrealistic.

There is no magical economic arrangement in which everyone sells premium-priced equipment forever while customers achieve unlimited returns.

Eventually, somebody must earn enough money to justify the bill.

When investors begin questioning whether those returns will arrive, the entire financial ecosystem becomes vulnerable to reassessment.

That is what makes the current selloff worth examining.

It isn't simply a judgment about Anthropic.

It is a reconsideration of how much the entire AI opportunity should cost.

The Federal Reserve Has Entered the Chat, and Nobody Is Happy

As though anxiety over AI valuations weren't sufficient, investors are also confronting a less accommodating interest-rate environment.

Recent Federal Reserve commentary has reinforced concern that inflation remains troublesome and that monetary policy may need to remain restrictive, or become more restrictive, than investors had hoped.

This matters enormously for growth-oriented companies.

A company's value depends partly on the cash investors expect it to generate in future years. When interest rates rise, those distant earnings generally become less valuable in today's dollars, all else being equal.

It is a perfectly reasonable financial concept.

Naturally, Wall Street reacts to it like someone just announced that birthdays have been canceled.

High-growth companies are particularly vulnerable because much of their valuation may depend on profits expected years into the future. When borrowing costs increase and investors can obtain better returns from relatively safer alternatives, the price they are willing to pay for distant growth can decline.

This doesn't mean the underlying business suddenly deteriorated.

It means the economic price of waiting has changed.

Consider two companies. One produces substantial free cash flow today. The other promises remarkable profitability a decade from now, provided market conditions cooperate, competition remains manageable, and its business strategy unfolds as anticipated.

When money is cheap, investors may be unusually generous toward the second company.

When money becomes expensive, patience acquires a price tag.

And suddenly that boring company producing cash today begins looking considerably more attractive.

I have never understood why this surprises people.

Interest rates are not obscure financial trivia. They influence everything from corporate financing to consumer spending and asset valuations.

Yet whenever monetary conditions become less favorable, investors act as though the Federal Reserve has personally betrayed an agreement to make everybody wealthy.

The central bank's responsibility is not to protect the market value of speculative technology stocks.

That may be disappointing to shareholders, but disappointment is not a monetary-policy objective.

The Market Isn't Always Rational, and Neither Are the People Who Claim It Is

I have spent enough time observing financial markets to recognize a recurring pattern.

When stocks rise, investors attribute the movement to improving fundamentals, visionary management, technological innovation, and their own extraordinary intelligence.

When stocks fall, the explanations become considerably less flattering.

Suddenly the market is irrational. Short sellers are manipulating prices. Analysts are incompetent. Algorithms are destroying capitalism. Somebody at the Federal Reserve must be stopped.

It rarely occurs to investors that the original purchase price might have been unreasonable.

This is what makes financial markets such effective instruments for exposing human psychology.

We are remarkably skilled at constructing explanations that preserve our self-image.

If a stock I own rises 20%, I am tempted to believe I possess exceptional judgment. If it falls 20%, I can manufacture a dozen explanations that avoid examining whether I made a poor decision.

The truth is usually less comfortable.

Sometimes I was right for the wrong reason. Sometimes I was wrong despite having a reasonable thesis. Sometimes the underlying company performs well while its stock declines because expectations were excessive.

And sometimes the market is simply repricing uncertainty.

The problem is that uncertainty doesn't make for particularly exciting financial programming.

Television requires confident people making dramatic predictions.

Nobody wants to watch a panel of analysts admit that tomorrow's market movements depend on countless variables they cannot reliably forecast.

That would be accurate, but terribly inconvenient for advertising revenue.

Instead, we receive predictions delivered with theatrical confidence by individuals whose previous predictions have quietly disappeared into the financial equivalent of a witness protection program.

I have learned to treat excessive certainty as a warning sign.

Anyone who claims to know exactly where Anthropic, Nvidia, or the Nasdaq will trade next week should explain why their extraordinary powers haven't already made them independently wealthy enough to avoid discussing stocks on television.

A Falling Stock Is Not Automatically a Bargain

One of the most dangerous phrases in investing is that a stock has become cheap because its price has fallen.

Cheap compared with what?

Yesterday's price?

Last month's all-time high?

The number somebody on social media insists it will reach?

None of those comparisons tells me what the business is actually worth.

A stock trading at $90 may be an extraordinary bargain if its underlying economic value is $150. It may also be wildly overpriced if reasonable assumptions suggest a value closer to $40.

The market price alone doesn't answer that question.

This becomes particularly important with newly public companies because investors often have limited public-market trading history and substantial uncertainty surrounding future performance.

A company can experience explosive revenue growth while struggling to convert that growth into sustainable free cash flow. Its market opportunity may be enormous while its competitive advantages remain vulnerable. Management may possess impressive technical expertise without having demonstrated equally impressive capital allocation.

These are not reasons to dismiss Anthropic.

They are reasons to evaluate it seriously.

I want to understand how quickly revenue is growing, what that growth costs, how much capital must be reinvested, and whether the company can defend its margins against formidable competitors.

I want to examine operating losses, future capital requirements, potential dilution, customer concentration, and the economics of delivering AI services.

I also want to know whether the assumptions embedded in the stock price leave any reasonable margin for error.

Because if every favorable development has already been priced into a stock, even excellent performance can disappoint investors.

That is one of the cruelest realities of investing.

A company can achieve tremendous things and still fail to achieve the even more tremendous things its shareholders were expecting.

The stock market does not award trophies for technological importance.

It prices claims on future economic benefits.

And those claims can become absurdly expensive.

The Great AI Bubble Debate Is Asking the Wrong Question

Every time technology stocks sell off, the same argument erupts.

One side declares that artificial intelligence is a bubble destined for catastrophe. The other insists that AI represents the greatest technological revolution in human history and that anyone questioning valuations is an intellectually deficient relic.

I find both positions exhausting.

The possibility that AI will transform society does not eliminate the possibility that investors have overpaid for companies participating in that transformation.

Both can be true.

In fact, history suggests that revolutionary technologies and speculative excess frequently arrive together.

The excitement surrounding a genuinely important innovation attracts capital. The capital finances experimentation, infrastructure, and competition. Investors become increasingly enthusiastic as adoption accelerates. Valuations expand, expectations become extravagant, and eventually market participants begin pricing assets as though every optimistic scenario is virtually guaranteed.

Then reality introduces complications.

Competition intensifies. Costs increase. Customers negotiate. Financing becomes more expensive. Revenue growth moderates. Some businesses fail, while others survive and prosper.

The underlying technology continues changing the world.

And investors who paid unreasonable prices discover that being correct about the future does not necessarily mean being correct about an investment.

The internet didn't disappear after the dot-com crash.

Neither did the economic damage suffered by people who purchased overvalued internet stocks.

This is the distinction I wish more investors understood.

A bubble does not require the underlying innovation to be imaginary. It only requires asset prices to become disconnected from reasonable economic expectations.

Conversely, a sharp market decline does not prove that an entire industry is fraudulent.

Sometimes an expensive asset becomes less expensive.

Sometimes a speculative narrative collapses.

Sometimes both things happen simultaneously.

My responsibility as an investor isn't to select the loudest ideological camp.

It is to distinguish technological opportunity from financial overpayment.

That job is considerably harder than posting rocket emojis beneath a stock chart.

What Anthropic Actually Has Going for It

Despite my criticism of speculative enthusiasm, I don't believe Anthropic should be dismissed simply because its stock is experiencing volatility.

The company participates in a rapidly developing market with potentially significant long-term demand.

Its Claude family of AI systems competes in areas where businesses may place considerable value on sophisticated reasoning, programming assistance, workflow automation, and reliable enterprise applications.

If Anthropic can develop products customers genuinely depend upon, it may have opportunities to build durable commercial relationships and meaningful recurring revenue.

That possibility deserves serious consideration.

Enterprise customers generally care less about dramatic product announcements than consumers might assume. They want useful products, predictable costs, security, integration with existing systems, and measurable improvements in performance.

An AI company capable of delivering those benefits consistently can become deeply embedded in customer operations.

That creates potential economic value.

However, I am equally aware that technological capabilities can spread rapidly. Competitors improve their models. Open-source alternatives become more capable. Customers compare offerings and demand better pricing.

The existence of substantial demand doesn't guarantee that the industry's current leaders will capture extraordinary profits indefinitely.

Competition has an irritating habit of interfering with beautiful financial projections.

Anthropic also operates in an industry where computing requirements, research expenditures, regulatory developments, and infrastructure availability can materially influence business performance.

Its opportunities are meaningful.

So are its risks.

I can respect the company without worshiping the stock.

I can appreciate the technology without assuming every valuation is justified.

And I can believe artificial intelligence has an extraordinary future without pretending I know which shareholders will enjoy the best returns.

That isn't pessimism.

It's the difference between investing and joining a fan club.

What I'm Watching Before I Get Excited About Buying

The current selloff has me interested, but interest is not the same thing as conviction.

I am not interested in buying a company simply because financial television has identified it as today's victim.

I want evidence that its business fundamentals, competitive position, and valuation justify assuming the risk.

With Anthropic, that means paying particular attention to the relationship between revenue growth and the cost of providing its services. I want to see whether growing demand can translate into improving unit economics and eventually sustainable cash generation.

I am also watching the competitive landscape.

Artificial intelligence is not a comfortable industry in which a company can develop one successful product and spend the next twenty years collecting effortless profits. Innovation happens rapidly, switching costs vary by application, and customers have powerful incentives to compare alternatives.

Management's ability to allocate capital intelligently matters just as much as technological ambition.

I also care about the broader market environment.

When interest rates remain elevated and investors become less willing to finance distant growth, highly valued companies can experience further compression even if their operating performance remains respectable.

That doesn't mean I can predict the bottom.

I can't.

And neither can the overwhelming majority of people pretending otherwise.

What I can do is examine businesses, establish reasonable valuation assumptions, consider unfavorable scenarios, and decide whether the potential reward adequately compensates for the uncertainty.

There is nothing particularly glamorous about that approach.

It doesn't produce exciting screenshots or viral investment predictions.

It does, however, reduce the likelihood that I will confuse enthusiasm with analysis.

And in a market increasingly dominated by enormous technological promises, that distinction is worth preserving.

The Real Problem Is That Investors Want Extraordinary Returns Without Ordinary Risk

I sometimes wonder whether the greatest technological achievement of the modern era has been convincing investors that risk can be eliminated through sufficiently enthusiastic storytelling.

Everyone wants to own the next trillion-dollar company.

Nobody wants to experience the volatility, competition, disappointments, and uncertainty involved in discovering which company will actually succeed.

Investors want explosive growth without expensive capital requirements. They want enormous competitive advantages without meaningful competition. They want revolutionary innovation without unexpected setbacks.

And naturally, they want the stock price to move upward in a dignified, uninterrupted line.

Unfortunately, financial markets have not signed that agreement.

A company can be innovative and financially vulnerable. A market can be enormous and intensely competitive. An industry can grow rapidly while profits accumulate unevenly among its participants.

This is normal.

What isn't normal is the expectation that buying a promising stock entitles somebody to immediate financial gratification.

Anthropic's IPO illustrates the problem beautifully.

Investors celebrated the company's debut, watched its shares climb, and then encountered an unpleasant reminder that public markets are under no obligation to sustain a celebratory mood.

The company did not cease to exist overnight.

Its products did not become irrelevant.

The long-term opportunities associated with AI did not vanish during a premarket trading session.

What changed was investor sentiment, alongside the market's assessment of risk, expectations, and future returns.

If that distinction makes somebody uncomfortable, perhaps the stock market is not the appropriate location for their emergency savings.

My Conclusion: The Intelligence May Be Artificial, but the Greed Is Entirely Human

I keep returning to one uncomfortable observation.

Wall Street has invested extraordinary amounts of money into machines designed to think more effectively, yet the people financing those machines continue displaying the same emotional weaknesses that have characterized financial markets for centuries.

Fear. Greed. Vanity. Impatience. Overconfidence.

We invent increasingly sophisticated technology, then use it to construct more elaborate explanations for why the price of something should rise forever.

Anthropic's volatile public debut is not proof that artificial intelligence has failed. Nor is it proof that investors should rush into the stock simply because its price has retreated.

It is a reminder that technological progress and investment success are different things.

The future may belong to artificial intelligence.

That doesn't mean today's shareholders automatically get to name their price for it.

I expect AI to continue changing how businesses operate. I expect competition to intensify, investment to remain substantial, and economic benefits to emerge unevenly across the industry.

Some companies will become enormously profitable.

Others will discover that impressive technology is not a substitute for an economically sustainable business model.

And investors will continue making the same mistakes they have always made, except now they can ask a chatbot to explain why those mistakes were perfectly reasonable.

My approach remains uncomplicated.

I want strong businesses, sensible valuations, durable advantages, disciplined management, and enough financial flexibility to endure circumstances that don't unfold according to the marketing presentation.

I want investments supported by actual economic value rather than the assumption that somebody even more enthusiastic will eventually pay a higher price.

Most importantly, I want to remember that a company's ability to change the world does not grant me immunity from paying too much for its stock.

Because the stock market has a remarkable talent for humiliating people who mistake a compelling story for a guaranteed return.

And if Friday's technology selloff teaches us anything, it's that even the most sophisticated artificial intelligence on Earth cannot protect investors from the oldest programming error in human history.

The belief that because something is extraordinary, any price must be reasonable.

Wall Street doesn't need a more intelligent chatbot.

It needs a little more common sense.

Unfortunately, common sense doesn't have an IPO.

And if it did, investors would probably bid it up 40% on the opening day, dump it the following morning, and spend the rest of the week explaining why nobody saw the collapse coming.

Some things never change.


Disclaimer: This article represents market commentary and opinion, not individualized investment advice. Market figures refer to the November 20, 2026, premarket period and may change throughout the trading day. Investors should review company disclosures, financial statements, valuations, and their own risk tolerance before making investment decisions.

Comments