BN · rank #1 · 2026-09-04
NVDA
NASDAQ · $5.52T (USD)
The exact numbers the algorithm saw.
| Composite score Z-score blend of the factors below; drives the rank. | 1.29 |
|---|---|
| Forward revenue growth Consensus forward revenue growth. | +63.3% |
| Forward net margin Consensus forward net margin. | +64.6% |
| Net margin TTM Trailing twelve month net margin. | +63.7% |
| Margin expansion Forward minus trailing net margin (percentage points). | +0.9% |
| Forward PEG Forward P/E to growth. Below 1 is cheap for the growth. | 0.56 |
| Debt / FCF Net debt relative to free cash flow. Lower is safer. | 0.31× |
| Analyst upside Spread between the consensus 12m target and the current price. | +42.7% |
| Last EPS surprise Most recent reported EPS versus consensus. | +6% |
| Market cap (USD) | $5.52T |
The AI research card
Independent qualitative review of each pick before the order is placed. On the rare day the research service is unavailable, the paper book trades on the quant ranking alone and no card appears here.
Summary
NVIDIA designs and sells AI accelerator platforms (Hopper/Blackwell GPUs, CUDA stack, networking) that are the de facto standard for data center AI training and inference, generating ~$89B in quarterly data center revenue.
Rationale
The bottleneck_solver thesis is directly reinforced — NVDA holds ~80-85% AI accelerator share, has Blackwell GPUs sold out into 2027 via $279B in cumulative supply agreements, and controls ~60% of global CoWoS packaging capacity, making it the literal chokepoint in AI compute infrastructure with a 0.56 forward PEG against 63% revenue growth.
Material risks
- 1U.S.-China export controls on H200/Blackwell are an active, escalating revenue constraint — not a forward-looking worry — that could structurally remove China as an addressable market and force product downgrades, directly capping the revenue visibility that underpins the bottleneck thesis.
- 2Hyperscale cloud customers co-developing custom silicon (e.g., the MediaTek partnership NVIDIA itself joined) signals the early stages of in-house capability building that, over a 2-3 year horizon, could erode NVDA's sole-source position and pricing power in its largest customer segment.
AI verdict council
Each pick is reviewed independently by 3 models before any order. 3 of 3 voted to proceed.