NVIDIA Coverage — The Durability Curve

NVIDIA is the central position in the AI infrastructure thesis. Its earnings, supply chain moves, and product roadmap are the primary signals for every downstream thesis — photonics, power grid, optical interconnects, and memory. This page indexes all Durability Curve coverage of NVIDIA as it develops.

Subscribe to The Durability Curve on Substack — free weekly analysis on AI infrastructure, verification, and durable systems.

Browse all company coverage.

Coverage

Three Signals the Market Is Missing on NVDA — NVDA earnings May 20, 2026: three non-obvious signals the consensus is overlooking. H200 China clearance, Culper Research shorts, and the optical interconnect supply chain signal.

Four Questions Every NVDA Investor Should Ask Before May 20 — Every research report tells you what the author thinks. Almost none tell you what would prove them wrong. Four falsifiable questions that test the NVDA thesis.

The Goldman vs KeyBanc Tension on NVDA — Two analyst calls from the last week capture the bull/bear tension better than any consensus number. Goldman at $1,040 vs KeyBanc's cautious note on near-term execution.

The Five Layers of the AI Infrastructure Thesis and What NVDA Tests at Each One — NVDA May 20 is not just an earnings print. It tests the entire AI infrastructure thesis across five structural layers, from GPU compute to grid power.

The Five Laws of Durable Systems Applied to NVIDIA — Five days out from earnings, tested against the five vault laws: bottleneck migration, difficulty as load-bearing, architecture outlives content, instruments over theory, the targeting problem.

Why NVIDIA Spent $500M on a Glass Company — NVIDIA's $500M Corning partnership means optical interconnects are the next AI bottleneck. Three new factories, 10x capacity expansion, and a supply agreement that reshapes the photonics thesis.

NVIDIA Earnings May 20 The Falsification Framework — Headline numbers matter less than the structural signals: supply chain capital commitments, optical spend trajectory, and the three numbers that reveal whether the bottleneck is migrating.

The 3 Numbers That Matter More Than NVDA's $80 Billion Quarter — Three numbers hidden in NVDA earnings that matter more than the revenue headline: purchase commitments, supply chain capacity, and the optical interconnect spend ratio.

The 3 Bottlenecks That Don't Care About NVDA's Earnings — Three structural bottlenecks that will outlast any single earnings report: power grid constraints, optical interconnect capacity, and HBM memory supply.

Adjacent Infrastructure Coverage

AI's Hidden Tollbooth — Why HBM Memory Is the Next GPU Constraint

The 5 Percent GPU Utilization Problem Redefining AI Infrastructure

Core Thesis

NVIDIA sits at the intersection of Law I (Bottleneck Migration) and Law II (Difficulty Is Load-Bearing). The market focuses on revenue beats and guidance. The structural story is where value migrates as compute scales: from GPUs to interconnects, from silicon to photonics, from chips to power. Every NVDA article on this page tests or expresses one of these laws.

Key Falsifiers

A durable NVDA thesis breaks if: supply chain capital commitments plateau while revenue grows (Law I inverse), optical interconnect capacity expands faster than compute demand (bottleneck resolves), or custom ASICs capture meaningful share from merchant silicon (Law II erosion).

Subscribe to The Durability Curve on Substack for weekly AI infrastructure analysis.

Full research reports at harryfloyd.gumroad.com.