“On two occasions I have been asked [by members of Parliament], ‘Pray, Mr. Babbage, if you put into the machine the wrong figures, will the right answers come out?’ I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.”
– Charles Babbage, Passages from the Life of a Philosopher (1864)
Just Press the AI Button
Do you really need glue on your pizza to keep the cheese in place?
Surprisingly, this was suggested by Google’s AI Overview tool regarding pizza preparation.
When making a pizza: “You can also add about 1/8 cup of non-toxic glue to the sauce to give it more tackiness.”
This humorous notion originates from an 11-year-old comment on Reddit by an individual with a rather colorful username. A hint? It starts with the letter F.
For those who have tried to learn a new language as adults, grasping slang and humor often proves to be the toughest challenge. AI tools, it seems, share this difficulty too.
Even the simplest aspiring chef can recognize that the suggestion to add glue to pizza is clearly a joke—no sane person would take it seriously, not even a child.
Yet AI missed the punchline entirely and treated it as practical advice.
The allure of AI lies in its promise of being smarter, faster, and more efficient than humans. It suggests that, with enough graphics processing units (GPUs), people can simply let AI do their thinking, freeing them for more enjoyable pursuits—like watching TikTok videos.
Not in the mood to do your taxes? Not eager to write a paper on 5th-century monks? Curious about the top 10 biotech stocks with the greatest potential? Want to use AI to anticipate market trends?
Don’t overthink it. Just press the AI button.
Stitching Sentences
From what we observe, generative AI takes a prompt and creates a collage of search engine results, stitching sentences together with common transitions. The result often lacks depth.
Have you noticed that Amazon now includes AI-generated summaries at the top of its customer reviews? They typically begin with “Customers say.” This content offers a formulaic summary.
Initially, it states what consumers appreciate about the product, followed by examples of features and benefits. This is often concluded with another sentence about additional points customers like, leading into a section on general complaints with the phrase “that said.”
The output resembles an incoherent mix of nouns, verbs, and adjectives, giving it a synthetic and artificial feel.
Amazon introduced these AI summaries to tackle what it terms review fatigue. Unfortunately, these AI-generated summaries often fall short. Why rely on an artificial product when you can easily read authentic reviews from real customers?
The example of glue on pizza underscores the limitations of AI-generated content. Presently, AI struggles to comprehend context, satire, and nuance, undermining the quality and accuracy of its results.
It is likely that AI will increasingly derive its content from other AI outputs. Are we ready for a barrage of nonsensical chatter? Expect a convoluted mix of gibberish devoid of the charm found in Lewis Carroll’s work.
So, where does the value lie?
The Future’s So Bright
The perceived value of AI is not solely about its current capabilities but rather the potential it holds for the future.
Bull markets thrive on compelling narratives, particularly those focused on emerging technologies.
Innovations like automobiles, telephones, cinema, radio, and aviation were the driving forces behind the bull market of the 1920s. The dot-com boom similarly captivated investors in the late-1990s.
While these technologies proved transformative, many pioneering companies did not survive their early successes. Even enduring firms, like Radio Corporation of America (RCA), experienced such volatility that many investors faced significant losses.
Have you heard of the Pierce Arrow Motor Car Company? On January 12, 1928, its President, Myron E. Forbes, confidently declared that, “There will be no interruption of our permanent prosperity.”
Sadly, within two years, that promised prosperity came to an abrupt end. By the decade’s close, the luxury car maker was bankrupt, and its assets were sold off.
Just like the roaring twenties and the late-1990s tech bubble, many investors falsely believed they were on the crest of wealth—driving up stock prices far beyond what earnings could substantiate. Eventually, these bubbles burst, leading to severe bear markets.
The excitement around AI has created a vibrant narrative, fueling bull market fervor. The last 21 months since the launch of OpenAI’s ChatGPT have indeed been thrilling.
Real technology giants like Microsoft, Google, and Apple have joined the race, eager not to miss out, dismissing the quality of AI applications since they believe, with enough GPUs, improvement will come in time.
How to Catch a Falling Knife
Here at Economic Prism, we’re not dismissing AI outright. Whether we embrace it or not, AI applications are swiftly becoming commonplace. We believe their true value lies in specific applications, not in crafting machines that replicate human thought.
It’s crucial for investors to truly understand what they are investing in. Since the introduction of ChatGPT, many have recklessly put their hard-earned money at risk, enchanted by the allure of quick riches.
NVIDIA, a leading supplier of advanced GPUs for AI processing, became the face of the AI bull market. Investors were relentless. On June 18, NVIDIA’s market capitalization reached $3.3 trillion. If NVIDIA were a country, it would rank as the sixth-largest economy globally.
However, the technology sector revealed signs of cooling over the last seven weeks. As of the market close on August 8, NVIDIA’s market cap stood at $2.6 trillion, indicating a staggering loss of about $700 billion in just two months.
Likewise, NVIDIA’s stock price, which peaked at $135.58 on June 18, subsequently fell to $104.97—a decline of 22.5 percent. Should you take the opportunity to buy the dip?
While we recognize NVIDIA as an exceptional technology firm, the pressing question is: at what share price would it be a worthwhile investment?
Like Buffett, we aim to acquire outstanding companies at reasonable prices. Currently, the narrative surrounding AI seems to have outpaced its genuine reality. Simply put, AI alone doesn’t justify NVIDIA’s peak valuation of $3.3 trillion.
Would $80 per share suffice? How about $60?
For us, these prices aren’t sufficient.
We’re looking for a decline of around 70 percent before we consider investing, which would bring NVIDIA’s share price down to approximately $40.
For context, accounting for the 10:1 stock split that took place on June 7, that price would position NVIDIA back to its value from October 31, 2023—Halloween.
At $40 per share, NVIDIA would have a market capitalization near $984 billion, substantially higher than Exxon Mobil’s market cap of $523 billion. Does that sound excessive?
Perhaps. Yet, at that valuation, assuming earnings remain relatively stable, we would be willing to cautiously average in as the share price hovers around $20.
In the meantime, if you dare, you can attempt to catch a falling knife. With a bit of luck, you might avoid getting hurt.
[Editor’s note: It’s remarkable how a few smart decisions can lead to significant wealth. Right now, I’m poised to make another smart choice. >> Let me show you how you can too.]
Sincerely,
MN Gordon
for Economic Prism
