- Griffin Asset Management
- Posts
- The End of the AI Rally?
The End of the AI Rally?
Market history offers useful context.

Hi All — Brian here.
Where were you on October 9, 2007? How about March 10, 2000? December 27, 2021?
These dates have one thing in common: they each marked a top for the markets, before brutal slides of as much as 81%. And like the “Black Monday” crash of October 28, 1929, they all occurred on seemingly unremarkable days.
Yet, in retrospect, the signs were there. Journalist John Cassidy, in his great book Dot.con: How America Lost Its Mind and Money in the Internet Era, documented that while the selloff began on a dull, uneventful day, it had been bubbling up well before the slide.
Soaring (or “nosebleed”) valuations. Televisions in bars and barbershops tuned in to Jim Cramer. Just about everyone you know asking if they should buy Bitcoin… or, pardon, Pets.com.
We’ve seen some of the same things recently. And, last month, investors reacted.
The Magnificent Seven stocks saw a nearly 5% pullback in November. Michael Burry, the trader famous for shorting the 2008 housing collapse, put $1 billion into a trade on the other side of the AI mania. Not even another earnings beat from chipmaking giant Nvidia, nor the release of Alphabet’s acclaimed Gemini 3 AI model, could reverse the bearish sentiment. The tech-heavy Nasdaq Composite finished the month in the red, down 1.5% in November.
There’s no doubt about it, bubble concerns are getting louder. However, it’s important to remember, dips can be healthy; a sign that markets are functioning as normal, not being overrun by excessive enthusiasm. They can also be buying opportunities, putting stocks on sale. And most notably, there’s a fine line between acting on these selloffs and trying to time the market, which no one can reliably do.
Case in point: Anyone who bought the Nasdaq in March 2000 (on the eve of its 81% slide) and hung on until today would be up 358%. It’s a reminder that time in the market beats timing the markets. Whether or not we’re in a bubble, that holds true.
NOVEMBER MARKET PERFORMANCE
S&P 500 | 6,849 | 0.1% |
Nasdaq | 23,366 | -1.5% |
Dow Jones | 47,716 | 0.3% |
NOVEMBER MARKET SUMMARY
The Bureau of Labor Statistics canceled October’s Consumer Price Inflation report and delayed the latest jobs data due to the lingering effects of the government shutdown.
The latest labor report from payroll processor ADP showed U.S. private employers shedding 32,000 jobs in November, a reversal from the 47,000 added in October, and the latest sign of a labor market slowdown.
Key Federal Reserve officials — including New York Fed President John Williams, considered the second-most influential U.S. central banker — supported another interest rate cut ahead of the central bank’s meeting on December 9-10.
Consumer confidence fell again in November, reaching its lowest point since the peak of tariff uncertainty in April. The Conference Board’s Expectations Index, which records Americans’ short-term economic sentiment, came in below 80 for the 10th straight month. Sustained readings below that threshold signal recessionary conditions.
ONE BIG THING: AI OVERVALUATION WORRIES
OpenAI’s chief financial officer Sarah Friar was perhaps too candid in a conference last month. The AI exec said the ChatGPT-maker might seek government assistance to finance and develop its models in the interest of growing the AI ecosystem “as fast as possible.”
These remarks prompted a round of handwringing, and comparisons to the government bailouts of 2008-2009 and the collapse of Lehman Brothers. Friar quickly took to LinkedIn to walk back her comments. Even so, the resulting selloff, which was further fueled by analysts’ overvaluation concerns, drove home the reality that expectations are very high for AI companies today. So high, in fact, that it may no longer be enough for earnings reports to be merely “excellent”.
Take Palantir Technologies, which reported earnings on November 4. Not only did its sales surge 63%, passing $1 billion for the first time, but net income more than tripled.
To all of that, Wall Street said, “Not good enough.”
Shares fell roughly 15% in the following days, as analysts digested the report. Later in November, the same thing happened to Nvidia. Its shares fell 8% in the wake of an earnings report, showing net income and revenue up 65% and 62% year-over-year.
These selloffs suggest that even outperformance from the biggest AI names is already priced in, and the faintest breeze in the other direction can put the rally on ice.
However, the actual numbers behind the valuation paint a somewhat rosier picture. The Nasdaq Composite today sports an average price-to-earnings ratio of 33.5, which is only marginally higher than its five-year average PE ratio of 31.5. However, both are quite a bit above its historical average of around 24.
So tech stocks do look expensive if you zoom out. But perhaps not prohibitively so.
Bulls tend to dismiss overvaluation concerns by pointing out that, with tech giants growing earnings by 65%, stocks will naturally trade at a premium. However, in doing so, they’re assuming such high growth can continue indefinitely. And even if it does, investors may remain unimpressed.
FEATURED POST
How to Mitigate Sequence-of-Returns Risk in Retirement
Sequence-of-returns risk describes the elevated impact of market volatility in the early years of retirement. And as market uncertainty abounds and concerns of inflated tech valuations grow louder, it’s an increasingly key consideration today for high-net-worth investors nearing their golden years.
KEEP READING
CLOSING REMARKS
Our team at Griffin Asset Management is here to help you make the most of the opportunities and challenges ahead.
If you have any questions or would like to discuss your financial strategy, please don’t hesitate to reach out.
We’re always available to set up a call and provide the personalized advice you need. Thank you for your continued trust and partnership.


