Buy Nvidia Today Because It's Likely To Be Worth A Lot More In 2 Year's Time

Nvidia reported $96.221B in quarterly revenue, with Data Center revenue of $89.023B up 117% year over year, and guided to $108.0B ±2% for Q3 as AI infrastructure demand remains supply constrained. Management expects approximately 70% fiscal-2028 revenue growth, while consensus fiscal-2028 EPS of $15.6826 has risen from $12.6737 over 90 days; the $327.70 average analyst target implies 41.9% upside from $230.86. Key risks are persistent constraints in power, memory, foundry capacity and labor through fiscal 2028, gross-margin pressure toward a 71%-72% floor, $279B in supply commitments, and potential reductions in AI-lab financing or hyperscaler capex.

Analysis

The relevant NVDA debate is no longer unit demand but conversion of booked AI infrastructure demand into cash earnings without a deterioration in working capital or gross margin. Extended receivables and vendor/lab financing can pull revenue recognition forward while shifting credit risk onto NVIDIA’s balance sheet; a sustained rise in DSO, receivables, or guarantees would deserve a lower earnings multiple even if reported revenue continues to beat. The cited forward-P/E figures also appear internally inconsistent across fiscal-year definitions, making calendarization of EPS and revenue estimates essential before treating the stock as optically cheap.

Over the next 1-3 months, the key catalyst is whether the next product transition preserves system-level pricing and margin while shipments scale. The more non-GPU bottlenecks constrain deployments, the more value accrues to memory suppliers, networking vendors, power/cooling and data-center infrastructure rather than solely to NVDA; conversely, delayed cluster commissioning can defer accelerator revenue despite nominal customer commitments. A margin outcome below the stated floor, or another material extension in cash conversion, would likely matter more to the share price than a modest revenue beat.

The contrarian risk is that consensus is extrapolating a supply-constrained market as though all constrained demand is economically durable. AI labs dependent on external funding and neoclouds reliant on GPU-backed financing are the marginal buyers; tightening private-credit conditions or weaker utilization would expose cancellation/renegotiation risk first in this cohort. Over 6-18 months, a successful transition to more interoperable custom silicon or lower-cost inference architectures could compress NVIDIA’s system ASPs even while total AI compute spending rises.

Key Decisions for Investors

• Maintain, but do not add aggressively to, NVDA ahead of the next earnings release; use a 3-6 month horizon and require confirmation that gross margin remains at or above management’s floor and DSO stabilizes. A margin miss or further material DSO expansion falsifies the near-term long thesis.
• For existing long exposure, buy a 3-6 month NVDA put spread financed partially with an out-of-the-money call sale only if implied volatility is below post-earnings historical ranges; the objective is protection against a multiple reset from financing/working-capital scrutiny rather than a bearish core view.
• Express the infrastructure bottleneck through a diversified long basket of SK Hynix, Micron (MU), Vertiv (VRT), Eaton (ETN) and Arista Networks (ANET) versus a smaller NVDA hedge over 6-12 months. The trade benefits if AI capex remains intact but spending mix broadens beyond accelerators; exit if hyperscaler capex guidance is cut broadly.
• Create an earnings watch item rather than a new directional position: verify customer concentration, receivables aging, inventory growth relative to shipments, guarantee obligations, and the portion of demand tied to financed AI labs. A meaningful deterioration in any two metrics should trigger a reduction in NVDA exposure regardless of a headline revenue beat.