Alphabet plans to double capex spending to a possible $185 billion—but it’s keeping CEO Sundar Pichai up at night

Alphabet announced a dramatic capex ramp to $175–185 billion in 2026 (versus $91.4B in 2025 and $52.5B in 2024), with Q4 capex at $27.9B as the company doubles down on AI infrastructure amid power, land and supply constraints. The company beat Q4 revenue and profit expectations, posted annual revenue above $400B and net income of $132.2B (up 15%), while services revenue rose 14% to $95.9B and YouTube topped $60B; Google Cloud backlog hit $240B (up 55% QoQ). Management says AI investments are already delivering results for search, ads and cloud, but flagged supply-constrained execution risks, prompting volatile after-hours stock moves and investor caution.

Analysis

Market structure: The capex arms race benefits hyperscalers and their ecosystem — GOOGL and MSFT gain persistent scale advantages (compute, models, distribution via the Apple deal) while chip suppliers and cloud infra (NVDA/TSMC, power/utility suppliers, select data‑centre REITs) see pricing power for 12–24 months as compute remains supply‑constrained. Incumbent enterprise SaaS (CRM, NOW) faces asymmetric downside: AI can displace layers of workflow automation, compressing renewals and upsell velocity over 6–18 months.

Risk assessment: Tail risks include antitrust/regulatory action, US export controls on advanced chips, or systemic power/grid bottlenecks that delay rollouts — any of which could convert capex into stranded assets. Expect immediate volatility (days–weeks) around guidance; margin pressure in next 2–4 quarters; and possible market share consolidation over 2–5 years if leaders monetize AI successfully. Hidden dependencies: TSMC capacity, regional grid upgrades, and top AI talent availability.

Trade implications: Prefer directional exposure to GOOGL (scale + distribution) and MSFT (enterprise AI adoption) while trimming SaaS exposure (CRM, NOW). Use options to time convexity: buy 12–18 month LEAP calls on GOOGL (~1–2% notional) and implement pair trades (long GOOGL, short CRM) to express relative thesis with stop losses and bounded size. Rebalance credit exposure: reduce long-duration IG tech credit if capex-driven issuance accelerates.

Contrarian angles: The market underestimates the Apple distribution multiplier for Gemini — this could accelerate monetization in 6–12 months and is partially priced out. Conversely, the capex surge may be partially overdone and create overcapacity in 2027–28; avoid full‑conviction long positions without staged scaling and explicit capacity/utilization read‑throughs.

Key Decisions for Investors

• Establish a 2–3% net long position in GOOGL (GOOGL/GOOG) sized to portfolio risk over 6–12 months; use a 10% stop‑loss and take profits at +20% unless fundamentals accelerate.
• Initiate a pair trade: long GOOGL (1.0%) vs short CRM (0.6%) for 6–12 months to capture AI monetization vs incumbent disruption; size to limit P&L to 1–2% of portfolio and reassess after two quarterly results.
• Buy 12–18 month LEAP calls on GOOGL equal to 1% notional (delta ~0.4–0.6) to capture asymmetric upside from Apple distribution and cloud backlog; finance by selling 90–120 day OTM calls (covered or naked sized conservatively) to harvest near‑term IV.
• Reduce exposure to SaaS leaders (CRM, NOW) by 30–50% over the next 60–90 days and redeploy into semiconductors (NVDA/AMD/TSMC) or MSFT given direct AI demand; re‑evaluate if CRM/NOW report AI monetization beats in two consecutive quarters.
• Reduce portfolio duration: trim long‑dated IG corporate bonds by ~20% of tech exposure within 3 months to hedge higher capex‑driven issuance and inflation; rotate into 2–5 year TIPS or short corporate paper until capex cadence and yield signals clear.