Q1 2026 Hyperscaler Earnings Land Into a Supply-Chain Crisis: $750B Announced Capex Meets a Helium Shortage and a 158 Percent DRAM Price Spike
Microsoft, Alphabet, Amazon, and Meta report Q1 2026 results today into a market where DRAM has repriced 158 percent year-over-year, Ras Laffan helium has been offline since February 28, and Asian chipmakers' buffer is expected to deplete between mid-July and mid-September. Alphabet 2026 free cash flow is projected down 90 percent; Amazon's is projected negative $17B. The announced AI capex story collided with the realized cost of building it.
Earnings Day Meets the Supply-Chain Reality
Microsoft, Alphabet, Amazon, and Meta all report Q1 2026 results today into a market that has spent three months absorbing the news that helium production fell off a cliff and DRAM costs more than doubled. The four-firm bloc reporting today, plus Oracle on its August fiscal calendar, has collectively committed somewhere between $660B and $750B of 2026 capital expenditure — approximately 75 percent of it AI-related — in what is the largest concentrated capital deployment into a single technology category in the history of the public markets.
That story has been the dominant AI-infrastructure narrative since late 2025. Today is the first earnings cycle that will surface, even partially, the question that supply-chain analysts have been asking since February: are the announced numbers still the right numbers given what it now costs to build a gigawatt of AI capacity?
The Two Blades
The top blade is the announced capex. Amazon's $200B 2026 figure represents 141 percent year-over-year growth. Alphabet sits at $175-185B, up from $75B in 2025. Microsoft is tracking toward $120B or more. Meta has guided to $115-135B, a 198 percent increase. Oracle adds $50B on its calendar. The five-firm aggregate sits between $660B and $750B depending on the analyst rollup; CreditSights' most recent estimate is approximately $750B.
The bottom blade is the supply chain. On February 28, 2026, Iranian missile strikes on QatarEnergy's Ras Laffan Industrial City complex disabled one of the two plants in the world capable of producing semiconductor-grade helium-4. Roughly 30 percent of global helium supply disappeared in 48 hours. Helium-4 has no synthetic substitute and the Algerian alternate plant runs at approximately 90 percent of nameplate in normal operation, with no realistic ramp to compensate. Liquid helium has a 35-to-48 day shelf life before evaporation losses make stored inventory unusable.
The combined effect has repriced AI compute inputs across the board. DRAM moved from $3.76 per gigabyte in 2025 to a forecast $9.71 per gigabyte in 2026 — a 158 percent increase. HBM3e and HBM4 — the high-bandwidth memory inside H100, H200, and B200 GPUs — moved proportionally with allocation rationing becoming the binding constraint above price for most enterprise purchasers. Nvidia H200 GPU spot pricing moved from $2.27 per hour in early January to $3.82 per hour by late April. Helium prices have run 40 to 100 percent above pre-strike levels.
The Free Cash Flow Squeeze
The clearest place to see the squeeze is in projected 2026 free cash flow versus announced capex for each of the four firms reporting today. Per Morgan Stanley:
- Amazon: $35B FCF in 2025 → projected −$17B in 2026. A $52B swing.
- Alphabet: $73.3B → projected $8.2B. A nearly 90 percent contraction.
- Microsoft: $62B → projected $21B. A two-thirds contraction.
- Meta: $52B → projected $12B. A 77 percent contraction.
None of the four firms is in financial distress; all generate enormous operating cash flow that funds the capex without external debt issuance. But the gap between operating cash flow and free cash flow is now the dominant story in the four hyperscaler income statements. The capex is not coming from the cash flow trajectory; it is consuming the cash flow trajectory.
The supply-chain pressure intensifies the squeeze in a specific way. Every dollar of helium-or-memory price increase is a dollar that comes out of FCF without producing additional capacity. The announced $750B aggregate buys less compute, less memory, and less inference capacity in late April than the same announced number bought in early February when the budgets were locked in. The hyperscalers are spending the same dollars and getting fewer chips.
Why This Cycle Is Structurally Different
Three features make the 2026 cycle structurally unlike past chip shortages. First, helium has no substitute and is not synthesizable at industrial scale — the supply curve is not elastic on any timeline shorter than several years of new well development. Second, AI demand is not consumer-cyclical; it is being driven by multi-year contracted commitments hyperscalers signed against specific customer demand. The demand curve does not retreat in a recession. Third, the customer concentration is unusually high — five firms competing for approximately 70 percent of global DRAM and a comparable share of leading-edge GPU capacity, none coordinated, each rationally pulling supply forward when allocation pressure rises.
The combination produces inelastic supply on a critical input, non-cyclical demand, and concentrated buyers each pulling in. The exit from this market structure is either a Hormuz resolution that brings Ras Laffan back online quickly, a major substitute-material breakthrough, or a demand-side adjustment by the hyperscalers themselves. None is likely on the quarters-not-years timeline the earnings models assume.
The H2 2026 Inflection
The single forecasting question that matters for the rest of 2026 is when Asian chipmakers' two-to-three-month helium buffer breaks. Holding variables constant at late-April values, the buffer exhaustion crossover lands somewhere between mid-July and mid-September 2026, with the consensus aggregator pointing at the second-to-third week of August plus or minus three weeks.
That window matters because it is the moment when supply-chain constraints become visible on earnings calls in language that current guidance does not yet permit. Q3 2026 earnings, reported in October-November, will be the first cycle where buffer exhaustion shows up as a delivered-capacity miss against committed buildouts.
Whether the hyperscalers preempt the moment by revising 2026 guidance downward in Q2 or hold the announced numbers through Q3 and surface the gap on the realized side is the strategic communication question that today's earnings calls are positioning against. Microsoft and Alphabet have institutional traditions of preemptive guidance management. Amazon and Meta have institutional traditions of letting the realization gap appear and explaining it after the fact. Today's calls reveal which strategy prevails.
What to Watch
The earnings prints land between 4:00 and 4:30 PM Eastern today. The three signals that will tell us whether the scissors hold their April 29 geometry, widen further, or partially close:
- The Q2 capex revision pattern. If two or more firms guide downward on 2026 capex within sixty days, the squeeze is being acknowledged at the public-disclosure level. If all four hold, the realization gap prints in Q3.
- The Hormuz traffic pattern. Strait reopening before mid-July shifts the buffer-exhaustion timeline right by a quarter or two. Continued contestation through August lands the inflection on schedule.
- The open-source frontier-model trajectory. If DeepSeek V4 or its successors close the production-agentic capability gap by Q3 2026, proprietary labs lose pricing power exactly when their input costs are still rising.
For deeper analysis of how the announced-versus-realized capex gap interacts with the April 2026 agentic foundation model pricing reset and the strategic consequences for engineering organizations, see our full analysis of the capex-supply-chain scissors.