The AI Monetization Split: AMZN Soars 15%, AAPL Slides 7% as Big Tech Earnings Diverge

The week ending July 31, 2026 will be remembered for what it proved and what it cost. On the proof side: artificial intelligence is generating real revenue at industrial scale, and the market rewarded the companies that can show it. On the cost side: the capital expenditure behind that revenue is staggering, the supply chain that feeds hardware demand is fraying, and the Federal Reserve — holding rates at 3.5–3.75% with three dissents and a 9–3 vote — made clear that the next move could be upward, not down. Fed funds futures now price a 64% probability of a rate hike in September.^[1] Friday’s July payrolls report, due August 7, is the next data point that will either reinforce or undermine that pricing.

The AI Monetization Divide

The cleanest read on the week is the spread between hyperscalers that monetized AI and those that spent on it without showing the return yet.

Amazon delivered the standout quarter. Net sales rose 20% year-over-year to $200.6 billion, and operating income surged 43% to $27.5 billion.^[2] The engine was AWS, which grew 37% — its fastest pace in 18 quarters — reaching a $169 billion annualized run rate. AWS’s AI and chip businesses each surpassed $25 billion in annual run rate.^[2] CEO Andy Jassy framed AI and core cloud services as driving each other’s growth. Amazon also raised its capital spending plan by roughly $20 billion, bringing the 2026 total toward $220 billion — and the market didn’t flinch. AMZN closed Friday at $271.58, up 15.3% on the day.^[3]

Microsoft reported its fiscal fourth quarter on July 29, posting $90.0 billion in revenue (+18% YoY) and $40.6 billion in operating income (+18%). Azure crossed $100 billion in annual revenue, growing 43% in constant currency. Copilot paid seats jumped to 30 million.^[4] The company guided to $255–260 billion in capital spending for fiscal 2027, and CFO Amy Hood said Microsoft should remain cash-flow positive.^[4] MSFT closed Friday at $464.72, up 3.0% on the day and roughly 16% for the week.^[3]

The contrast with Apple was stark. Apple posted a record June quarter — $109.4 billion in revenue, up 16% year-over-year, with iPhone sales surging 22%.^[5] But the company then issued September-quarter guidance below consensus, citing supply constraints: chip shortages and rising memory costs that had already forced price increases in June.^[5] AAPL closed Friday at $308.91, down 7.4%.^[3] It was also Tim Cook’s farewell earnings call; his successor, John Ternus, takes over in September.^[5]

Meta showed the cost side of the AI buildout most viscerally. Revenue edged past estimates at $60.8 billion, but EPS missed at $6.18 as net income fell 14% year-over-year and expenses surged 55%. Free cash flow collapsed to $784 million.^[6] Reality Labs posted a $4.62 billion operating loss.^[6] The stock tumbled 9.5% in after-hours trading on July 29 before recovering through the week to close Friday at $556.71, up 3.3%.^[3]

Google (Alphabet) also delivered a strong print, closing Friday at $356.13, up 6.7%.^[3]

Friday’s Scorecard

All prices as of 16:00 ET close, July 31, 2026.^[3]

The Fed: A Hawkish Hold With Teeth

On July 29, the FOMC voted 9–3 to hold the federal funds rate in a range of 3.5%–3.75%.^[7] The three dissents came from Cleveland’s Beth Hammack, Minneapolis’s Neel Kashkari, and Dallas’s Lorie Logan — all arguing for tighter policy with inflation above the Fed’s 2% target for more than five years.^[7] Chairman Kevin Warsh, in only his second meeting, has pushed for giving the markets fewer forward signals about the central bank’s next move.^[7]

The macro backdrop explains the tension. The June FRED snapshot showed CPI inflation at 3.46% year-over-year — nowhere near the 2% target — with the 10-year Treasury yield at 4.68% and the 2s10s curve steepened to +0.47%.^[8] Unemployment sits at 4.2%, and real GDP grew 2.1% year-over-year. Consumer sentiment, however, plunged to 49.5 — down 18.4% from a year ago — a divergence from the headline growth figures that the Fed cannot ignore.^[8]

Fed funds futures now assign a 64% probability to a rate hike in September.^[1] That is not a consensus view; it is a market that read three hawkish dissents and an inflation overshoot and concluded the next move is more likely up than down. The 40% — the case for holding or cutting — rests on consumer sentiment at recession-adjacent levels, a weak June payrolls print of just 57,000, and the possibility that growth is slowing beneath the surface.

What the VIX and Breadth Are Saying

VIX closed the FRED snapshot period at 20.66, up 29% year-over-year but up 17% month-over-month.^[8] That is not a panic reading, but the month-over-month acceleration is consistent with a market that knows it is at a pivot point.

The weekly index performance tells a story of narrow leadership:

• S&P 500: 7,489.72 (+1.05%)^[1]
• Nasdaq Composite: 25,373.85 (+1.59%)^[1]
• Dow Jones: 52,485 (+1.04%)^[1]
• Russell 2000: 2,931.34 (+0.05%)^[1]

The Russell 2000 barely moved — small caps gained less than a tenth of a percent on the week^[1] — while the mega-cap-driven indices posted solid gains. Friday’s ETF tape confirmed the same pattern: SPY +0.72%, QQQ +0.65%, DIA +0.54%, but IWM −0.48%.^[3] Energy (XLE +1.00%) was the only sector ETF to outperform, while financials (XLF −0.11%) and broad tech (XLK −0.22%) lagged.^[3] The market is being carried by a handful of hyperscaler names, and the breadth underneath is thin.

The Historical Analog Worth Noting

The FRED macro snapshot’s nearest historical analogs cluster around mid-2006 and October 2007.^[8] In both periods, unemployment was near 4.6–4.7%, inflation ran in the mid-3% to low-4% range, and the Fed held rates at an elevated level. The October 2007 analog is particularly uncomfortable: the economy was one month from entering recession, the Fed had stopped hiking but had not yet begun cutting, and the yield curve had only recently un-inverted. None of this means a recession is imminent in 2026 — real GDP at 2.1% and unemployment at 4.2% are meaningfully better than 2007’s deteriorating profile — but the analog is a reminder that “the Fed holds and inflation stays sticky” is a pattern that has preceded trouble before.

What to Watch Next

1. July Nonfarm Payrolls — Friday, August 7, 8:30 AM ET. This is the single most important data point for the September FOMC. June’s print of 57,000 was a shock; economists expect a rebound to roughly 125,000 for July, with the unemployment rate potentially ticking up to 4.3%.^[9] A number at or below 57,000 would strengthen the case that the economy is slowing — and complicate the hawkish-hold narrative. A strong print above 150,000 would cement the 64% hike probability.

2. Apple’s supply chain in the September quarter. The guidance miss was attributed to chip and memory shortages, not demand weakness.^[5] Whether those constraints ease or worsen through the iPhone launch cycle will shape the Q4 outlook for the entire semiconductor supply chain.

3. Meta’s free cash flow trajectory. With FCF down to $784 million and expenses up 55%, the question is whether AI infrastructure spending is a one-time ramp or a permanent cost structure.^[6] Meta’s next quarter will show whether the AI investment is generating the ad-revenue uplift Zuckerberg described.

4. The capex reckoning across hyperscalers. Amazon raised 2026 capex to ~$220 billion.^[2] Microsoft guided to $255–260 billion for FY2027.^[4] At what point does the market start discounting the gap between capital spending and the revenue it generates?

5. FOMC speakers and the September meeting. Chairman Warsh’s preference for fewer forward signals means the data — not guidance — will do the talking. Watch every FOMC speaker appearance between now and September for whether the hawkish dissenters gain or lose ground.

The trajectory is clear: AI monetization is separating winners from losers, and the Fed is leaning hawkish into an economy where the surface numbers look solid but consumer confidence is deteriorating. The 60/40 skew toward a September hike could flip on a single payrolls print. That is where the real uncertainty lives.