GOOGL Downgraded: Capex Surge Threatens Alphabet Margins
Alphabet faces a fresh downgrade as $700B in AI capex across Big Tech raises depreciation and margin compression risks heading into Q2 earnings July 28.
Key Points
- Alphabet has been downgraded on Wall Street with analysts citing AI capital expenditure risks, margin compression, and accelerating depreciation that could weigh on earnings through 2027.
- Alphabet, Amazon, Meta, and Microsoft are collectively projected to spend more than $700 billion on AI infrastructure in 2026, a capex cycle with no historical precedent and uncertain return timelines.
- GOOGL reports Q2 2026 earnings on July 28 — that print is the first hard data point on whether the AI spend is driving revenue growth fast enough to offset the cost surge.
Alphabet has been downgraded by at least one major Wall Street firm, with analysts zeroing in on a risk that the market has been willing to overlook for most of 2026: the four largest AI spenders — Alphabet, Amazon, Meta, and Microsoft — are collectively on track to deploy more than $700 billion in capital expenditures this year, and the depreciation clock on that infrastructure starts running the moment the assets go live. For GOOGL, which last reported a 90.39% EPS beat on $5.11 versus the $2.68 estimate, the question is no longer whether the AI buildout is working — it's whether the cost curve is about to catch the revenue curve.
The Capex Math That's Spooking Analysts
The $700 billion collective AI capex figure is not a projection range — it is a baseline estimate built on announced spending commitments from each of the four companies individually. Alphabet alone has signaled capital expenditure plans that dwarf anything in its history. The company's infrastructure investments span data centers, custom TPU silicon, subsea cables, and the physical real estate to house the compute required to train and serve next-generation models. Each of those asset classes depreciates on a different schedule — servers typically over three to five years, buildings over 20 to 40 years — but the near-term income statement impact lands in the same place: higher depreciation charges flowing through cost of revenues and operating expenses.
The depreciation risk is not hypothetical. When a company builds $20 billion worth of data center capacity in a single year and places those assets into service, the straight-line depreciation on that spend hits the next 12 quarterly income statements. For a company with Alphabet's revenue scale — trailing 12-month revenue above $350 billion — that is manageable if cloud and advertising revenue grows faster than the depreciation drag. The problem analysts are now flagging is the timing mismatch. The capex is being deployed today, the assets are going into service across 2026 and 2027, but the AI-driven revenue upside — new products, higher-margin cloud contracts, monetized Gemini integrations — is still largely a 2027 and 2028 story.
Alphabet's last reported EPS of $5.11 represented a 90.39% beat against the $2.68 consensus, which sounds like a business firing on all cylinders. But consensus estimates are a lagging indicator of analyst conviction, and the size of that beat partially reflects how badly the Street had marked down expectations after the 2024–2025 advertising softness. The forward question is whether the next beat, due July 28, is structurally supported or whether rising depreciation and elevated operating costs have started compressing the margins that drove the prior upside.
What a $700 Billion Capex Cycle Actually Means
To put the $700 billion collective AI infrastructure commitment in context: the entire U.S. interstate highway system, built over roughly 35 years and adjusted for inflation, cost approximately $530 billion in today's dollars. The four largest American technology companies are proposing to spend 32% more than that — in a single calendar year — on compute infrastructure whose primary output is probabilistic text generation, image synthesis, and autonomous software agents. That is either the greatest capital allocation decision in corporate history or the most expensive overshoot since the telecom fiber buildout of the late 1990s.
Alphabet joined a €411 million funding round for Proxima Fusion on July 7, 2026 — a fusion energy bet that signals the company is not limiting its infrastructure ambitions to conventional data center architecture. That investment won't move the needle on 2026 financials, but it illustrates a capital allocation philosophy that prioritizes long-duration moonshots over near-term return optimization. For income-oriented or value-conscious institutional holders, that philosophy is a feature. For traders positioned around the July 28 earnings print, it is a source of uncertainty because it makes the capex trajectory harder to forecast quarter to quarter.
The macro backdrop adds another layer of complexity. With the 10-Year Treasury yield at 4.48% and core CPI at 2.8% year-over-year, the cost of capital has normalized at levels meaningfully above the near-zero regime that made unlimited AI investment feel essentially free. Alphabet carries net cash, so it is not borrowing to fund its capex — but the opportunity cost of deploying tens of billions in cash into infrastructure rather than buybacks or dividends is real and rising as yields stay elevated. The Fed Funds Rate at 3.63% means the hurdle rate for any capital project has to clear that threshold on a risk-adjusted basis before the first line of code generates a dollar of AI revenue.
What Traders Need to Watch Before July 28
The GOOGL downgrade lands at an uncomfortable moment in the calendar. Q2 2026 earnings are scheduled for July 28 — exactly 20 days away — and the print will be the first hard data point on whether the AI infrastructure buildout is translating into revenue acceleration at Google Cloud, which is the business unit most directly tied to the capex story. Google Cloud grew at 28% year-over-year in its most recent reported quarter. Analysts need to see that number hold above 25% and ideally accelerate toward 30% to justify the current capex commitment in the eyes of the market.
Alphabet's most recent 8-K filing provides the baseline regulatory disclosure framework traders should monitor for any updated guidance language or material event disclosures ahead of the earnings call. Watch specifically for any segment-level commentary on Cloud backlog — that metric, more than any single quarterly revenue number, tells you whether enterprise customers are committing to Alphabet's AI stack at the volumes needed to justify the infrastructure investment.
The technical picture for GOOGL is also worth tracking. The stock is trading in a range that reflects the tug-of-war between the strong Q1 EPS beat — which sent shares higher in April — and the growing analyst concern about capex-driven margin compression in the back half of 2026. A clean downside break below the April pre-earnings level would be a signal that institutional holders are de-risking ahead of July 28 rather than holding through the print. Conversely, any pre-earnings analyst note that quantifies the depreciation impact — putting a specific dollar figure on the 2026 and 2027 drag — would give the market something to price against, which tends to reduce uncertainty-driven selling and potentially set up a relief rally if the actual Q2 numbers come in above the newly reset bar. The July 28 earnings call is now the single most important near-term event for GOOGL holders, and the downgrade issued this week has effectively moved the goalposts — a 90% EPS beat will no longer be sufficient if the capex guidance language suggests the cost curve is steepening faster than the revenue curve.
The Weekly Investor
Daily market analysis for active traders. Free.
