Everything About AI
NextFin News - The Federal Reserve held interest rates steady Wednesday in a split 10-2 vote, and hours later Alphabet reported third-quarter revenue that fell short of Wall Street's expectations on weaker advertising growth, forcing investors to confront two uncomfortable questions at once: how much higher rates must go to kill inflation, and how much AI spending the companies funding it can survive before the cash flows break. One day delivered two messages, and both were about the price of money.
The central bank left its benchmark rate in the 3.75% to 4% range, with Fed Governor Lisa Cook and Cleveland Fed President Beth Hammack dissenting in favor of a quarter-point hike. Minutes after the 2 p.m. ET announcement, the S&P 500 and Nasdaq each rose about 0.3%, and Treasury yields fell once Fed Chairman Kevin Warsh began his news conference. Then, after the closing bell, Alphabet said sales excluding partner payouts came to $111.2 billion, just below the $111.3 billion analysts had projected, and net income of $2.12 a share missed the roughly $3 consensus estimate. Shares fell about 8% in after-hours trading, erasing a regular-session gain of 1.34% that had lifted the stock to $324.89.
The Fed said it needs more data before it moves again. Alphabet showed that even a cloud business roughly doubling in size cannot fully offset a slowing advertising engine. The common thread is cash: the cost of money is staying high, and the cost of building AI is only getting higher. Investors who expected a clean either-or day, rates or earnings, got both, and the market's verdict on each was the same: show me the cash.
The Fed's Split Vote Is a Warning, Not a Pause
The 10-2 vote was not the widest dissent of this tightening cycle, but it was a clear signal that the committee's center of gravity has shifted. Hammack also dissented in July, when the committee held rates in a 9-3 vote with two other officials joining her. The split matters less for the decision itself, which was widely expected, than for what it signals about the committee's tolerance for patience: a central bank divided on whether it has already done enough is a central bank more likely to overshoot.
The statement itself leaned hawkish. Officials upgraded their assessment of the economy, saying "economic activity is expanding at a strong pace," a step up from the "solid pace" cited in September, while adding that inflation "remains elevated." The committee kept the door open without committing to a date: "The accumulation of more information will help the Committee determine the policy stance appropriate to ensure a timely return of inflation to the Committee's 2 percent goal," policymakers said. The Fed raised rates in September for the first time since 2023, in a unanimous 12-0 vote, and Wednesday's hold was framed as a pause for data, not a pivot.
Warsh did not claim victory in his press conference. "What we've done in this monetary medicine is starting to work some, but there's no firm conclusions," he said. "I'm encouraged by the progress we've made in a short period of time. But if the crux of your question is 'you haven't delivered yet, you haven't achieved price stability,' you're right about that. But we're working on it."
The data leaving the Fed behind it is not the data of a disinflation victory lap. Consumer prices rose 3.7% in the 12 months through September, with gasoline alone jumping 9.1% in the month, the Labor Department reported. U.S. crude settled at $99.90 a barrel on Tuesday, and oil topped $115 during Wednesday's session. Energy is the transmission belt: pricier crude lifts gasoline and shipping costs, which eventually show up on grocery shelves, and those pass-throughs are exactly what keep core inflation sticky.
The bond market made its own assessment. The 10-year Treasury yield, the benchmark that mortgage rates follow, closed Tuesday at 4.99% after peaking at 5.31% on Oct. 5, and added 5 basis points on Wednesday to 5.047%. The two-year yield, most sensitive to Fed policy, rose 3.9 basis points to 4.776%. When a central bank holds rates but the long end of the curve refuses to cooperate, the market is effectively tightening policy for it. That is the mechanism by which a "pause" still squeezes borrowers: the term premium, the extra yield investors demand for holding long-duration risk, is a fear tax on the future, and it is being levied right now.
Markets took the hint on the near-term path. Traders priced in an 86% chance of a quarter-point hike at the Fed's next meeting on Dec. 9, according to CME FedWatch, and expect the funds rate to climb from about 3.9% now to nearly 4.7% by September 2027. That expectation barely moved on the day, which is its own kind of message: the decision was priced, the bias was priced, and the only thing left to price is how far the cycle goes.
The second-order effect is the part investors are not fully pricing. A "higher for longer with a hawkish bias" stance does not just raise the discount rate on future earnings; it raises the hurdle for every capital-intensive business plan that depends on cheap money. When the risk-free rate pays nearly 5%, the market stops rewarding ambition and starts demanding receipts. That is the bridge between the Fed's meeting room and Alphabet's earnings call: both are about whether the future cash flows promised today are worth what they cost to build.
Alphabet's Miss Is About the Old Business, Not the New One
Alphabet's quarter was not a disaster on its face. Total revenue of $127.6 billion rose 25% over the year-earlier quarter, marking a 13th consecutive quarter of double-digit gains, but slightly undershot the $127.8 billion consensus forecast. Google Cloud brought in $30.71 billion, beating the $29.1 billion analysts had projected, with growth of roughly 100% year over year. The advertising business, however, delivered $82.01 billion versus expectations of $83.5 billion, and YouTube's ad revenue grew 9% to $11.2 billion, short of the $11.43 billion estimate.
That mix is why the stock fell about 8% after hours despite the cloud beat. Cloud is the growth story investors came for; advertising is the cash cow that pays for it. When the cow slows, the capex story comes under scrutiny no matter how fast the cloud business is growing. The market's math is unforgiving: a cloud business growing at triple-digit rates is worth a premium only if the advertising engine that funds it keeps producing the cash to build it.
The scrutiny is specific, and it is about cash. Alphabet's free cash flow turned negative for a second straight quarter, falling to -$9.06 billion in the third quarter from -$5.85 billion in the second. In the second quarter alone, the company spent roughly $44.9 billion on capital expenditures, and it has guided 2026 spending to between $195 billion and $205 billion, up from a prior range of $180 billion to $190 billion. Money is being burned faster than the advertising engine can replenish it, and the company repurchased no shares for a second straight quarter, against $13.2 billion of buybacks a year earlier.
CEO Sundar Pichai defended the spending on the earnings call, pointing to the company's latest frontier AI model, Gemini 4 Carbon, which he said outpaces Gemini 4 Argon and beats competing models including OpenAI's GPT-6 Astra and Anthropic's Claude Opus 5.5. He also said the Gemini app has more than 1 billion users, making it the fastest-growing product in Google's history, and noted the company is on track to record half a trillion dollars in total annual revenue. On the competitive front, Oracle said Tuesday it will be the first company to purchase Google's AI processors for its own data centers, while Meta's new Muse AI agent has already been set up by 10 million people, according to CEO Mark Zuckerberg.
The question the market is now asking is whether the advertising slowdown is cyclical or structural. A cyclical answer means budget timing, election-year caution, and a re-acceleration next year. A structural answer means AI search agents are eating queries, and every query migrated to an agentic answer is a query that may never see an ad. The quarter gave evidence for both: cloud demand is real and accelerating, but the ad business, which still funds the whole machine, is losing momentum. YouTube's 9% ad growth, while solid, is a far cry from the double-digit rates that funded the last cycle of investment.
The Two Stories Are One Story
Read together, the Fed decision and the Alphabet miss describe the same squeeze from two directions. On one side, the cost of capital is staying high and probably rising. On the other, the AI build-out is turning into a capex supercycle that shows few signs of peaking. Companies that need to borrow to fund their AI ambitions now face both a higher price for money and a market that is losing patience with promises of future returns.
This is a cyclical pressure layered on top of a structural shift, and the distinction matters. The Fed's pause is cyclical: it is data-dependent, election-adjacent, and reversible the moment inflation prints hot again. The AI build-out is structural: the rules of the industry have changed, and no major player can afford to stop spending without ceding the next platform. But the market's tolerance for that spending is cyclical, and it is tightening. The build-out will not stop, but the valuation premium for it can and probably will compress. That is the call this market is making, quietly, in the spread between cloud revenue and stock price.
The clearest evidence that the market is repricing patience came before the bell. Oracle jumped 12% after issuing fiscal 2030 revenue projections of at least $225 billion, implying annualized growth of more than 35% from fiscal 2026. Investors cheered a company promising to turn AI infrastructure into contracted revenue. Alphabet, by contrast, was punished for spending heavily while its core business slowed. The market is not against AI; it is against AI without a visible path to cash. Microsoft, reporting the same evening, faced the same test, and the divergence between the two companies' stock paths since July, when Microsoft's cloud growth hit a four-year high, shows that investors are grading each dollar of capex on its own return, not on the sector's promise.
The cross-asset transmission runs deeper than tech. A 5% risk-free rate changes the denominator for every asset class at once. Long-duration growth stocks lose value mechanically as future cash flows are discounted more heavily. Leveraged borrowers, from AI data-center developers to commercial real estate owners, face refinancing walls at rates that did not exist when they made their plans. Even the dollar is caught in the squeeze: foreign capital has poured into U.S. mega-cap equities and hyperscaler bonds chasing the AI trade, and if that narrative stumbles, the dollar's safe-haven role could collide with its new role as a risk-sensitive bet on American exceptionalism. One rate path, many breaking points.
What Could Break This Read
The strongest counter-thesis is that the Fed is simply buying time and that inflation is already rolling over on its own, making the December hike a non-event and relieving pressure on growth stocks. That view has a name, the soft-landing consensus, and it is what has kept equities resilient through two rate hikes. It is also the view that a single hot inflation print could overturn, because the Fed has explicitly tied its hands to incoming data.
The falsifying signal is quantifiable. If core PCE prints at 0.3% month over month or higher for two consecutive months, the "pause is enough" thesis is wrong and the market's 86% December-hike probability will look too low, not too high. On the Alphabet side, if advertising revenue re-accelerates above 10% year over year next quarter while cloud growth holds above 50%, the structural-search-decline thesis is wrong, and this quarter will be read as a one-quarter pause rather than a trend change. Until one of those prints, the market will keep pricing both risks.
"You haven't delivered yet, you haven't achieved price stability," Fed Chairman Kevin Warsh said Wednesday, acknowledging that the job of containing inflation is unfinished even as the central bank held rates steady in a divided vote.
"Search usage and growth is back in focus given Muse launch and challenges Google could face in migrating users to agentic while maintaining strong monetization," a research note ahead of the results warned, capturing the market's anxiety about Google's advertising core.
What to Watch Next
The next Fed meeting is Dec. 9, six weeks away, and the committee's preferred inflation gauge, the core Personal Consumption Expenditures Price Index, is due Thursday with projections above 3% year over year. For Alphabet, the key signals are Q4 advertising guidance, whether free cash flow returns to positive, and whether cloud growth can hold its pace against a rising comparison base. For the broader market, the watchlist is the yield curve: if the 10-year pushes back toward its 5.31% October peak, the squeeze on long-duration assets tightens regardless of what the Fed says.
By time horizon, the picture splits. In the short term, sentiment and liquidity rule, and both the Fed's hawkish bias and Alphabet's cash burn argue for volatility in long-duration tech. Over the medium term, fundamentals decide, and the companies that convert AI demand into contracted revenue, not just usage, will separate from the pack. Over the long term, the structural shift is real: AI infrastructure spending is not a cycle that reverts, it is a new cost of doing business, and the winners will be the ones whose cash flows survive the transition.
Three scenarios frame the path from here. The base case, and the one markets are pricing: the Fed hikes once more in December, inflation grinds down slowly, and tech volatility persists as capex discipline separates winners from the rest. The upside case: core PCE rolls over faster than expected, the Fed pauses through year-end, and Alphabet's advertising business re-accelerates, restoring the cash flows that fund the AI build-out. The downside case: oil stays above $100, inflation prints hot twice in a row, the December hike becomes two, and the 10-year yield retests 5.3% while ad budgets keep weakening. In that world, the companies with the most ambitious capex plans and the least free cash flow have the furthest to fall.
Who benefits: short-duration value stocks, cash generators, and energy names tied to oil's rise. Who is exposed: long-duration, unprofitable AI infrastructure names and any company funding a capex build-out with debt. The week's lesson is that in a world of 5% risk-free returns, ambition without cash flow is not a strategy; it is a bet that the Fed will blink first.
The Fed held rates, but it did not reassure. Alphabet grew cloud, but it did not satisfy. The market's verdict on both was the same: show me the cash.
Explore more exclusive insights at nextfin.ai.
