From Marx's vampires to Buffett's moats, the crumbling Gothic castle has been capitalism's favorite metaphor for 250 years.
Fortune
3h ago
Ai Focus
As the AI transaction exposed the cracks in the "noble mansion" of early 21st-century capitalism, the metaphors of "castles" and "moats" long used on Wall Street have once again come into focus. Tech stocks such as Alphabet, ServiceNow, and Intel have recently seen significant declines, prompting the market to re-evaluate whether the moats surrounding these AI giants can truly hold them off.
Helpful
No.Help

The market is welcoming a particularly Gothic Halloween: The AI trades have exposed cracks in that crumbling aristocratic mansion of early 21st-century capitalism. For years, Wall Street has liked to describe itself using terms like "castles."

Jamie Dimon has long talked to JPMorgan Chase shareholders about the bank's “fortified balance sheet.” In his 2015 letter to shareholders, he dedicated a section to explaining what this term meant. By 2021, he added a second term, “castle,” warning that “the moat protecting this company is not particularly deep.”

The term "moat" originates from Warren Buffett. In an article co-written with Carol Loomis in 1999, titled "Fortune", Buffett argued that the key to investing lies in a company's competitive advantage and whether this advantage is sustainable, and that companies worth holding should be surrounded by a wide moat.

One of the favorite dangerous scenes in Gothic novels is a castle that is on the verge of collapse. For about 250 years, these two elements have been interchanging imagery: castles and moats, vampires, and creatures created in laboratories. This Halloween season, they have once again come to the forefront of discussion; on Thursday and Friday, tech stocks plummeted for two days, with the focus on whether the moats of the AI giants could hold firm, while Gothic horror is also becoming popular on bookshelves and in cinemas.

According to The Numbers data, horror films accounted for 15.5% of the U.S. domestic box office in 2025; Circana BookScan data shows that sales of horror books in the United States increased by 7% in the first nine months of 2025, while overall sales of adult novels only increased by 1%. The New York Times asked on Friday: Why has every book become gothic in style nowadays? Publishers attach this label to a variety of works ranging from family legends to space operas; several film critics believe that the highly popular commercial scandal documentary You Can See Everything, which is Nathan Fielder, is a romantically twisted story that can even be considered gothic in nature, continuing the tradition of Alfred Hitchcock.

The moat is being put to the test.

Selling pressure started from Alphabet. The stock fell 12.6% on Thursday, and its market value evaporated by about $500 billion after the parent company of Google announced its quarterly results. Core search business revenue fell short of expectations, partly due to the popularity of " AI mode " searches, which have lower profit margins, as well as the company spending money faster than anticipated. FactSet data shows that its free cash flow was negative $9.1 billion, while analysts were expecting it to be negative $2.23 billion.

The worst performer in the S&P 500 index was the software giant ServiceNow. On that day, the stock fell by 27% due to slower revenue growth and disappointing guidance. By Friday morning, it was Amazon and Intel that came under pressure, with Intel falling by about 25% just before the market opened. According to The Financial Times, as AI spending surged again in the third quarter, the total market value of large tech companies evaporated by nearly $900 billion. On Friday alone, the market value of the Nasdaq 100 index components decreased by over $1 trillion. Reddit fell by more than 22% before the market opened on Friday, and the company is tightening its policies: as AI dominates search, it is shifting towards a post-login experience.

What worries investors is the irony of it: AI is eroding those businesses that generate cash, yet that cash is then reinvested into AI. According to S& P Global Market Intelligence's data, hyperscale cloud providers have raised $326 billion through stocks and bonds this year, which is more than three times the amount spent in 2025.

Apollo Chief Economist Torsten Slok raised questions about the underlying reasons on the morning of the Alphabet decline to his clients. He wrote in his Daily Spark notes on Thursday: "As the potential overthrow of AI seems imminent, there is a lot of discussion currently about moats (a metaphor for competitive advantages)."

He continued the line of thinking of the finance professor Aswath Damodaran – this “godfather of valuation.” Damodaran compiled a list of sources of competitive advantage, including brand strength, switching costs, network effects, cost advantages, and legal protection. Subsequently, Slok simplified it into a AI moat checklist, posing three questions: Does the company possess data that competitors do not have? Has its product become an integral part of customers’ daily work? Can it raise prices without causing customers to leave?

From the third test, it seems that the company selling AI is not qualified. The new GPT-6.1 model of OpenAI has a price for each token that is only half of the previous version's price. The index compiled by the fee management company Ramp shows that, based on the actual prices paid by enterprises to Anthropic and OpenAI, the price of token has decreased by about 35% since the end of June, which Axios refers to as a price war.

The company that paid for AI has not yet received much in return. In a note dated October 24th, Slok caused a stir on Wall Street, stating that AI “has not yet brought any financial returns to American businesses, except in the tech industry.” He pointed out that while each employee’s income and profit margin of the “Big Seven” are increasing, the corresponding indicators for the rest of the S&P 500 companies remain largely unchanged; mainstream companies have begun to regard token expenditures as cost centers that drag down profit margins, rather than sources of growth.

The first castle is fake.

A researcher at the Manchester Gothic Studies Center, Xavier Aldana Reyes, told The Times that the popularity of Gothic literature often rises during times of social anxiety and political turmoil. Literary agent Mary C put it more bluntly, saying to the newspaper, "Gothic works are meant to expose the decay at the core of those things that seem glamorous or wealthy."

This literary genre was invented by a person who came from a wealthy background, was familiar with the old order from within, and was able to receive a salary without working at all. Horace Walpole was the son of the British Prime Minister Robert Walpole. His work " The Castle of Otranto ", published in 1764, is generally regarded as the first gothic novel.

He disguised this book as a translation of a little-known Italian work, claiming that it was discovered in the library of an ancient Catholic family in northern England and printed in Naples in 1529. However, according to a 2014 paper by ankaya from Ertu Ü rul Ko ç and Neslihan Atcan Altan, this is actually a dramatized representation of the process by which the noble system was replaced by the “greedy” emerging bourgeoisie. Literary historian Robert Miles writes that the modern novel itself is considered to have been radically transformed by printing technology, which also made it possible to have circulating libraries, which explains that set of pre-modern pseudo-origin stories.

We are all familiar with the subsequent Gothic story. Mary Shelley mentioned in the preface she wrote for Frankenstein in 1831 that this idea originated in June 1816, when she overheard Lord Byron and Percy Shelley discussing experiments with electric current. Live Science made a direct analogy in an article reprinted in 2018: the debates sparked by electricity back then are exactly identical to those sparked by AI today.

Karl Marx formalized this concept of exchange. In his 1867 work "Das Kapital," he wrote that capital is "dead labor that lives only by sucking the living labor," a metaphor that reappears throughout his writings over a period of 30 years. In 1897, Bram Stoker created " Dracula ," which shaped the great myth of the modern vampire; at that time, the wealth and estates of the British nobility were threatened by agricultural depression, while cheap agricultural products from prosperous American farms also flooded the market.

The literary critic Franco Moretti interpreted the count as a businessman. In an article published in New Left Review in 1982, he described the vampire in Dracula as a rational entrepreneur: he invested in gold, expanded his sphere of influence, and conquered the City of London.

The Immortal Things on Wall Street

In more recent times, an article published in Rolling Stone in 2009 by Matt Taibbi titled The Great American Bubble Machine referred to Goldman Sachs as "a giant vampire squid entwined around the face of humanity." This statement spread so widely that Goldman Sachs' then CEO admitted in August of the same year that the article "hit a nerve" in some respects.

Economists also have their own "monsters": zombie companies. Researchers Ryan Banerjee and Boris Hofmann from the Bank for International Settlements define these as companies that have been established for at least 10 years and have not been able to cover their interest expenses for three consecutive years. In 14 developed economies, the proportion of zombie non-financial companies listed on stock markets rose from 4% at the end of the 1980s to 15% in 2017. The researchers found that weak banks allow zombie companies to continue to linger by extending loans rather than writing them off, and lower interest rates tend to increase the proportion of zombie companies.

The latest candidates come from the software industry. According to PitchBook statistics, the scale of leveraged loans with prices below 80 cents face value has reached $131 billion, which is the highest level since May 2020 and also represents a common threshold for distress. Software companies account for 32% of these distressed loans, yet they only hold a 12% weight in the index; moreover, 84% of the software companies under pressure are backed by private equity. These are like castles that cannot afford to pay their own maintenance costs and can only stand on the strength of other people's money, akin to "dollar princesses" who are arranged with credit lines rather than a dowry.

This kind of anxiety is becoming a big business in the "haunted prosperity" of 2026. Publisher Zando has launched a horror-themed sub-brand Evil Twin, planning to release 6 to 8 books each year. In Hollywood, Warner Bros.' R-rated body horror film Clayface won the North American weekend box office championship with approximately $24 million on the weekend of October 23rd, according to The Numbers data.

These days, machines are also creating such content. According to the novel writing platform Inkfluence AI, in the 180 days up to September 12th, 3,929 horror novels were started on the platform, with haunted houses and ghost themes being the most popular, accounting for about one-third of them in total.

This article utilized the generative AI for data retrieval and/or transcription. The journalist independently reported and verified the factual statements in this article, and manual editing was conducted before publication. This article has been updated to include the trends on NASDAQ on Friday.

Tip
$0
Like
0
Save
0
Views 50
WalletJYS reminds readers to view blockchain rationally, stay aware of risks, and beware of virtual token issuance and speculation. All content on this site represents market information or related viewpoints only and does not constitute any form of investment advice. If you find sensitive content, please click“Report”,and we will handle it promptly。
Submit
Comment 0
Hot
Latest
No comments yet. Be the first!
Related
Tech stocks tumbled, dragging down the Nasdaq 100 for its third pullback this year
The Nasdaq 100 index tumbled 4.5% on Friday, triggered by poor financial reports from tech companies, falling more than 10% from its recent high during the session, marking the third pullback of 2026. Amazon reported negative free cash flow of $42.4 billion over the past 12 months, and Intel's revenue guidance for the fourth quarter also fell short of expectations.
Businessinsider
·2026-11-01 03:50:24
14
XRP Why is the "80% increase over history" in November a statistical trap?
The article states that the average increase of XRP in November seemingly reached as high as 80.2%, but this figure has been severely distorted by two abnormal surges in 2013 and 2024; a median of -10% is more in line with reality. The author also mentioned that XRP has fallen below the key support level of $1.30, and on November 1st, 1 billion XRP will be unlocked as planned. The short-term trend will depend on changes in supply, liquidation, and liquidity.
U.Today
·2026-11-01 03:40:33
21
Google's first data center satellite has been launched, and a haiku was also written
Google stated that its prototype space data center satellite Suncatcher has established communication with the ground since its launch earlier this month and has begun in-orbit testing. The satellite is equipped with a TPU and a pre-installed DeepMind Gemma model 4, and subsequent tests will focus on evaluating the chip's performance under thermal stress and space radiation conditions.
TechCrunch
·2026-11-01 03:30:00
23
AI Stocks Lost Billions of Dollars in October, While Ordinary U.S. Stocks Performed Significantly Better
In October, tech stocks in the U.S. market faced selling pressure, but most of the constituents in the S&P 500 actually rose in value. Since the S&P 500 is weighted by market capitalization, the decline of a few large tech stocks dragged down the index, while the equal-weight version rose by about 5% during the same period.
Coinpaper
·2026-11-01 03:02:32
27
OpenAI Cuts Financing Target by $10 Billion, AI Stock Sell-off Threatens a $1.25 Trillion Valuation
According to Reuters, citing the Financial Times, OpenAI has lowered its latest funding target from $30 billion to approximately $20 billion, and its valuation has also decreased from $1.4 trillion to $1.25 trillion. With the significant selling of artificial intelligence stocks, investors' willingness to fund the industry's massive infrastructure plans is waning.
Coinpaper
·2026-11-01 03:02:30
27
View More