The panelists agreed that while AI-related stocks offer attractive opportunities, there are significant risks to consider, such as the durability of AI moats, regulatory scrutiny, and geopolitical risks. They suggested selective bets on durable AI winners while being cautious about potential headwinds.
Risk: The durability of AI moats and potential regulatory/data constraints were the most frequently cited risks.
Opportunity: Selective bets on durable AI winners, notably Alphabet, were seen as attractive opportunities.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Key Points
- AI stocks now make up a large percentage of both the S&P 500 and Nasdaq-100.
- Following two pieces of advice from Warren Buffett could prepare investors if the market eventually crashes.
- These 10 stocks could mint the next wave of millionaires ›
Artificial intelligence (AI) stocks have helped lead the market higher …
Read more
Key Points
- AI stocks now make up a large percentage of both the S&P 500 and Nasdaq-100.
- Following two pieces of advice from Warren Buffett could prepare investors if the market eventually crashes.
- These 10 stocks could mint the next wave of millionaires ›
Artificial intelligence (AI) stocks have helped lead the market higher over the past few years. While that has helped power the S&P 500 (SNPINDEX: ^GSPC) and Nasdaq Composite (NASDAQINDEX: ^IXIC) to new all-time highs, it has also left both indexes very top-heavy with leading AI stocks.
For example, eight of the S&P 500's largest holdings are tech stocks that make up more than 35% of its holdings. Most of these are top semiconductor companies and hyperscalers (owners of large data centers). It's even more dramatic for the popular Nasdaq-100, whose 10 largest holdings are tech stocks and account for nearly half its portfolio.
Missed Nvidia in 2009? This Rare Signal Is Flashing Again. In 2009, a "Double Down" signal flashed for a little-known chipmaker called Nvidia. For the first time in years, that same "Total Conviction" signal is flashing for a company 1/100th the size of Nvidia. Continue »
At the same time, most of these stocks' fortunes are very tied to each other. Hyperscalers are spending a tremendous amount of money building out AI infrastructure, which, in turn, is driving growth at semiconductor stocks. If hyperscalers don't see a strong return on their AI investments and this spending stops, then these stocks could go down in unison, dragging down the major indexes with them.
One of investing stalwart Warren Buffett's favorite measures of whether stocks are overvalued is to take the total value of the U.S. stock market and divide it by gross domestic product (GDP). The former CEO of Berkshire Hathaway has said a range of between 75% and 90% is reasonable, while stocks start to look overvalued when it rises above 120%. This measurement, sometimes called the Buffett indicator, is currently over 235%, signaling a very expensive market.
Other popular metrics, like the cyclically adjusted price-to-earnings ratio, or CAPE ratio, also point to a frothy market. This metric smooths out earnings by adjusting for economic cycles, seasonality, and inflation over the past 10 years. Historically, this index's long-term average is around 17.4. It rose above 42 in August and is at its highest level since before the dot-com crash in 2000.
Now, whether the market has formed an AI bubble and whether the stock market will crash anytime soon is uncertain. However, with these indicators pointing to a frothy market, it is best to be prepared. Let's look at two things Buffett would recommend.
Have cash ready
One of Buffett's most famous pieces of advice is to "Be fearful when others are greedy, and greedy when others are fearful." Before he stepped down as CEO at Berkshire Hathaway at the start of this year, he had been following his own advice, not chasing stocks and instead letting his cash holdings climb.
One of the best things an investor can do before a potential market crash is to have some cash on the sidelines. This does not mean sell all your stocks and sit in cash. That would be a mistake, as no investor or market indicator can 100% predict when a market crash will occur.
However, if the market has a major pullback and you have no cash to invest, that also doesn't help. As such, if you want to follow Buffett, don't be fully invested and have some cash ready to pounce.
Buy companies with durable competitive advantages
Another important lesson investors can take from Buffett is which types of stocks to buy during a big market pullback. When the dot-com bubble burst, there were some great stocks to buy, but also many that never made it.
In an article in Fortune magazine, Buffett described the type of stocks investors should be looking for, saying, "The key to investing is not assessing how much an industry is going to affect society, or how much it will grow, but rather determining the competitive advantage of any given company and, above all, the durability of that advantage."
In other words, not all AI stocks will be long-term winners. Look for companies with wide moats that give them a long-term competitive advantage. Buffett seemed to leave one hint on an AI stock to buy if the market crashes, with his last big bet being on Alphabet (NASDAQ: GOOGL) (NASDAQ: GOOG), which is the most complete AI stock with both world-class chips and AI models to go along with a strong distribution and ad network moat.
Where to invest $1,000 right now
When our analyst team has a stock tip, it can pay to listen. After all, Stock Advisor’s total average return is 978%* — a market-crushing outperformance compared to 213% for the S&P 500.
They just revealed what they believe are the 10 best stocks for investors to buy right now, available when you join Stock Advisor.
**Stock Advisor returns as of September 2, 2026. *
Geoffrey Seiler has positions in Alphabet. The Motley Fool has positions in and recommends Alphabet and Berkshire Hathaway. The Motley Fool has a disclosure policy.
AI Talk Show
Four leading AI models discuss this article
Opening Takes
“Deploy incrementally into durable AI leaders like Alphabet, as the AI upgrade remains a secular driver of earnings and value, even amid volatility and valuations.”
Buffett-esque caution about cash and moats is sensible, but the article understates two risks. First, the AI capex cycle could persist well beyond a few quarters, keeping hyperscalers hungry for GPUs and data-center bandwidth and supporting continued upside for AI leaders. Second, the durability of a moat in AI-linked businesses is evolving; Alphabet-type platforms benefit from data networks and ad moats, yet regulatory scrutiny, competition from open AI–style models, and possible ad-market shifts could compress returns. Inflation and cash drag also argue against simply waiting in cash. Bottom line: selective bets on durable AI winners, notably Alphabet, remain attractive despite headline froth.
The strongest counter: if rates stay high and AI-driven growth slows, valuations on AI leaders could compress quickly; the moat thesis may fail as models commoditize and regulatory risk rises.
“AI-driven index concentration creates real downside risk, but structural margin and rate shifts may keep valuations elevated longer than the Buffett indicator alone implies.”
The article correctly flags dangerous concentration, with eight tech names over 35% of the S&P 500 and ten names near 50% of the Nasdaq-100, plus correlated capex between hyperscalers and semis. Yet it underplays that today's Buffett indicator above 235% and CAPE near 42 partly reflect permanently higher margins, lower rates, and cloud revenue already flowing, not pure speculation. Buffett's own Alphabet purchase shows he sees durable moats inside AI rather than avoiding the sector entirely. Cash reserves help, but the real test is whether ROI data from Microsoft, Google, and Amazon capex turns negative in 2025-26.
Even structurally higher margins cannot prevent a rapid de-rating if one hyperscaler reports disappointing AI returns, triggering simultaneous selling across semis and indexes that broad indicators failed to predict in 2000.
“Aggregate valuation metrics like the Buffett Indicator are becoming obsolete due to the fundamental shift toward high-margin, capital-efficient tech dominance in the S&P 500.”
The article's reliance on the 'Buffett Indicator' is increasingly dated. While the 235% market-cap-to-GDP ratio suggests extreme froth, it ignores the structural shift in the S&P 500 toward asset-light, high-margin software and hyperscale businesses that generate massive free cash flow compared to the industrial-heavy economy of the 20th century. Betting on a crash based on these macro metrics misses the reality that AI capital expenditures are currently being funded by balance sheets with historically low net debt. Investors should focus less on the aggregate index valuation and more on the widening delta between the 'Magnificent Seven's' cash-generative moats and the rest of the index's stagnant earnings growth.
If interest rates remain 'higher for longer,' the high-multiple valuations of AI hyperscalers will face a brutal discount rate adjustment that renders their current cash-flow-to-Capex efficiency irrelevant.
“Concentration risk is real and dangerous, but the article conflates it with a valuation bubble that may not exist if AI capex ROI continues to materialize.”
The article conflates two separate problems: concentration risk (real) and valuation (overstated). Yes, the Mag 7 dominates the S&P 500 and Nasdaq-100—that's structural risk if capex ROI disappoints. But the Buffett Indicator at 235% is a blunt instrument; it ignores that U.S. GDP excludes trillions in intangible asset value that AI companies are creating. The CAPE ratio at 42 is genuinely elevated, yet the article doesn't acknowledge that AI semiconductors and cloud infrastructure have higher structural margins than pre-2000 dot-com peers. The real risk isn't valuation per se—it's correlation. If hyperscalers' AI ROI disappoints, NVDA, MSFT, GOOGL, and TSLA all crater together, dragging the index. But that's a 12-18 month thesis, not imminent.
If AI capex ROI is actually materializing faster than consensus expects—and early LLM monetization data suggests it might—then current valuations aren't bubble-level, just expensive. The article assumes capex stops; it could instead accelerate profitably.
The Debate
Responding to Gemini
“AI moats may prove less durable than current cash-flow expectations, risking a rapid re-rating if capex ROI slows.”
Gemini's point about Buffett indicators misses a key risk: AI moats may not stay durable as models commoditize and regulatory/data constraints bite. Even with cash-rich balance sheets, capex ROI could deteriorate if monetization lags, open-source competitors erode differentiation, or ads/data rules compress moat economics. If ROI slows or capex intensity relaxes, the Magnificent Seven could re-rate quickly despite today’s cash flow, a scenario the piece downplays.
Responding to ChatGPT
“Regulatory and capex risks could combine to trigger a 2025 selloff in AI leaders despite structural margins.”
ChatGPT overlooks how sustained high capex from hyperscalers could mask weakening ROI for quarters, delaying any re-rating. Claude's correlation thesis gains force when paired with potential ad-market compression from privacy rules, creating a synchronized selloff across GOOGL, META, and NVDA suppliers. This linkage between regulatory and capex risks remains under-discussed, raising the probability of a 2025 drawdown even if margins hold structurally.
Responding to Grok
“The AI rally ignores a massive geopolitical discount related to supply chain concentration in Taiwan that could trigger a systemic collapse regardless of software ROI.”
Grok and ChatGPT are missing the geopolitical tail risk: the 'AI moat' is increasingly dependent on TSMC’s precarious position in the Taiwan Strait. If we are discussing capex ROI and correlation, we must acknowledge that a single supply-chain shock renders hyperscaler balance sheets irrelevant. The current valuation ignores this 'geopolitical discount' entirely. Even if ROI materializes, the concentration of physical manufacturing risk in one region makes the entire AI-driven rally fragile, regardless of software-side margins or cash flow.
Responding to Gemini
“Taiwan supply risk is real but a secondary trigger; capex ROI disappointment in 2025 is the primary catalyst for re-rating.”
Gemini's Taiwan risk is real but overstated as *the* valuation driver. TSMC concentration matters for capex execution timelines, not current ROI math. The actual sequencing risk: hyperscalers report disappointing AI monetization in Q2-Q3 2025 *before* any geopolitical shock materializes. Regulatory ad-market compression (Grok's point) hits faster than Taiwan Strait escalation. Geopolitical tail risk deserves a discount, but it shouldn't eclipse the nearer-term capex ROI deterioration thesis.
Panel Verdict
NEUTRAL No ConsensusThe panelists agreed that while AI-related stocks offer attractive opportunities, there are significant risks to consider, such as the durability of AI moats, regulatory scrutiny, and geopolitical risks. They suggested selective bets on durable AI winners while being cautious about potential headwinds.
Selective bets on durable AI winners, notably Alphabet, were seen as attractive opportunities.
The durability of AI moats and potential regulatory/data constraints were the most frequently cited risks.
Related News
Nvidia Is Likely to Remain in an Uptrend as Earnings Outlook Remains Bright
AMD Stock Alert: Advanced Micro Devices Teams Up With Cisco on AI Infrastructure
VGT’s $58 Billion AI Chip Concentration Could Derail Your 2027 Returns
History Says September Is Brutal for Stocks. Is This the Year AI Stocks Prove It Wrong?
This is not financial advice. Always do your own research.