Tom Lee Sees the Market as a Loaded Spring That Three Triggers Could Release

Last week showed once again how dependent markets have become on the artificial intelligence trade. A single round of headlines about OpenAI revenue first triggered sharp selling, then a fast rebound the next morning. Bloomberg reported that the company sat near 50 billion dollars in annualized revenue at the end of September, with a target above 70 billion by year end. Schwab wrote in its October 9 morning note that this clarification, plus steadier oil prices, lifted technology names early. Thursday had seen losses of 5 to 8 percent across names like CoreWeave, ARM and Intel. Friday flipped the picture.

A Spring Coiled Toward Fear

Fundstrat partner Tom Lee offered the striking diagnosis. He described equities as a spring coiled toward fear rather than euphoria. Yahoo reported that Lee has met investors on the road since September 28 and found almost everyone skeptical of the AI boom. Many read the picture as an outright bubble. Fundstrat research notes from early October use the same concept and stress that the wall of worry stands unusually high. Pessimism runs so deep that good news meets suspicion and earnings beats get sold. The tighter the coil, the smaller the spark that moves prices hard.

The first release valve, in Lee view, is third quarter earnings. His logic is plain. Whether AI spending earns its keep cannot be settled with words, only with numbers. Yahoo quoted him saying third quarter results would restore confidence in AI. Large technology firms built enormous computing capacity and must now show it converting into revenue. Strong figures would weaken the bubble story and pull sidelined buyers back in. Weak figures would release the spring downward. So this is no ordinary earnings season, it is judgment day for the thesis.

The second trigger sits in politics. Lee argues the midterm elections work as a clearing event. Once voting ends, one of the uncertainty items markets hate disappears. Which party takes the Senate matters less than the removal of the question mark. He expects political pressure on AI names to ease through the same channel. The claim can sound odd at first, since chip demand and ballot boxes share no direct link. But risk appetite is a collective mood, and as uncertainty lifts investors grow readier to bet on the future.

Why Concentration Can Be Defended

Lee came prepared on the concentration critique. A few giant firms driving AI investment looks natural to him and signals widening moats. He noted that nearly all of them are American businesses. His analogy came from finance. Nobody calls banking overconcentrated while JPMorgan and Goldman grow enormous, yet the same scale draws criticism in AI. Bigness here means scale advantage rather than inefficiency. Concentration does not erase risk, since a stumble by a few names drags the index. But Lee judges a small set of efficient capital allocators healthier than a scattered crowd.

The third trigger is rates, where Lee also breaks with consensus. Rates are no longer a negative story for equities, in his reading. The case is technical and concrete. Commercial short positions against rates sit at records while the upward move has stretched far. Not everyone can borrow at these levels, so capital demand weakens. A Fundstrat note from October 7 had the 10 year touching 5.30 percent with the 30 year at 5.69. TradingEconomics data showed the 10 year hovering near 5.24 percent on October 9. Either we stand at the peak or relief sits near, with little room above. Even small easing would let valuations breathe.

Racing Bonds While Cash Waits Aside

High rates carry a side that works for stocks. Lee estimates elevated rates now generate roughly 10 trillion dollars a year in investment income for American households. That money will not all stay in bonds, some inevitably flows into equities. If rates fall, bond competition softens and multiples expand. If they hold, the income stream keeps feeding the market. Like water behind a dam, the stillness stores pressure rather than calm, and direction becomes everything. That mechanism is why Lee reads current levels as fuel for pent up demand rather than a threat.

The competition is real and should not be denied. Seven year investment grade corporate bonds yield above 6 percent, while municipal bonds top 7 percent for buyers in the highest tax bracket. Against dividend stocks paying 4 to 5 percent, those are serious alternatives for anyone seeking calm returns. ICI data showed money market fund assets rising 72 billion dollars to 7.96 trillion in the week ended October 7, a fresh record. Lee argues this money is not abandoning stocks, but the slice rotating into bonds slows any equity rally. Should yields ease, some parked cash would return toward risk assets.

The fate of AI capital spending stays the knot at the center. Microsoft investor materials show the scale. Third quarter capital expenditure reached 31.9 billion dollars, with two thirds going to short lived assets like graphics processors and server chips. Above 40 billion is expected next quarter. Lee's Microsoft example stands out. Copilot going model neutral, working with OpenAI or Anthropic alike, made the product stickier and opened return potential. But final judgment comes early for no one. Whether the spending was wasted or converts into profit will only be known in a couple of years.

Placed side by side, the three triggers clarify the picture. If earnings validate the thesis, confidence returns. If elections cut uncertainty, one overhang lifts. If rates ease, bond competition softens. Lee expects all three to land in the coming weeks and sees the spring releasing upward. The reverse stays possible. One OpenAI headline, one missed earnings print, or one fresh rate spike triggers selling. The market is not relaxed, it is compressed and waiting for confirmation. Like any loaded spring that can fire either way, numbers rather than fear will pick the direction.