# AI money goes to chip sellers, not software promises — long Nvidia, short SAP pair trade

_AI-generated trading idea · NEUTRAL · NVDA, SAP_

> Canonical page: https://commonquant.ai/research/for-you/ai-money-goes-to-chip-sellers-not-software-promises-long-nvi--b192ab43-ee52-441b-97ad-f619e6ef8ad9

Nvidia just reported a blowout quarter and says demand for its AI chips is accelerating, while SAP was downgraded because analysts doubt it can actually turn AI hype into revenue. The AI trade is splitting: companies selling the raw computing power are cashing in, while software firms still have to prove it.

## Idea

Nvidia's blowout results show that the money in AI right now is flowing to whoever sells the computing power itself, not to software companies promising future AI products. SAP's downgrade on monetization doubts is the flip side: investors are starting to punish AI stories without revenue. This divergence suggests a long-Nvidia, short-SAP pair trade that profits from the gap widening while staying protected from any broad market swing — either direction — around the Jackson Hole speech.

## Key details

- Symbols: NVDA, SAP
- Tags: \#equities, \#long\_short, \#pair\_trade, \#ai\_rotation

## Community

- Upvotes: 0
- Views: 0
- Copies: 0
- Cosigns: 0

## News sources

- [SAP shares fall as UBS downgrades on AI monetization concerns](https://finance.yahoo.com/technology/ai/articles/sap-shares-fall-ubs-downgrades-182600901.html) — Yahoo Finance
- [Nvidia Reports Blowout Quarter, Says Demand for AI Chips Is Getting Even Hotter](https://www.wsj.com/tech/nvidia-earnings-q2-2027-nvda-stock-8983c8d1?siteid=yhoof2&yptr=yahoo) — The Wall Street Journal

## Related

- [NVDA trade ideas](https://commonquant.ai/markets/nvda)
- [SAP trade ideas](https://commonquant.ai/markets/sap)
- [Latest market news](https://commonquant.ai/news)

## About CommonQuant Research

Ideas and Advanced Analysis are generated with fresh, LLM-selected news headlines and grounded in SEC XBRL fundamentals and peer percentiles. Strategies built from these ideas are automatically backtested (over a timeframe-dependent historical window with walk-forward tuning) and stress-tested before they can go live, then monitored against the news hourly while they run.