Dell reported on September 1 and jumped nearly 10% in extended trading; HPE reported a day later, finished September 3 up 5%, then gave back 4.5% on September 4. Both described record AI demand, but the market treated backlog quality, margins, and execution differently. Dell Technologies Inc. (NYSE:DELL) and Hewlett Packard Enterprise Company (NYSE:HPE) are selling the servers and networking around accelerators, not the accelerators themselves.
Dell booked $60.9 billion of AI-server orders, recognized $16.4 billion of AI-server revenue, and exited the quarter with a $95 billion backlog. The bull case is conversion at scale across more than 6,500 customers, with storage and services available for cross-selling. The bear case is mix. Large AI systems can carry thinner margins, require working capital, and concentrate the supply chain around scarce components.
Insider Monkey counted 77 hedge funds holding Dell Technologies Inc. at June 30, up from 72 at March 31. Cliff Asness’s AQR Capital Management disclosed 879,477 shares after reducing its position by 18%. The quarter-end filing predates the earnings surge and does not show whether AQR later changed course.
HPE posted record quarterly revenue of $12.2 billion, up 34% year over year, while its Cloud & AI business generated $9.0 billion and server revenue reached $6.8 billion. Its bull case is broader enterprise infrastructure, especially after adding Juniper networking. Its bear case is supply constraints, integration, and pressure from AI-system mix and pricing even when orders are strong.
Hedge-fund breadth for Hewlett Packard Enterprise Company rose to 85 funds in Q2 from 58 in Q1. Paul Singer’s Elliott Management reported 32,271,985 shares, 18% more than in the previous quarter.
The August 14 short-interest settlement recorded 59,572,842 HPE shares sold short, equal to about 4.5% of float, with 3.34 days to cover, but it predates both earnings reports. Dell’s larger backlog gives investors a clearer near-term revenue bridge, while HPE must prove that Juniper and financing turn demand into better margins. The split reaction is rational if investors reward visibility more than headline growth. For both, operating income and cash conversion matter more than the raw dollar value of AI orders. Backlog cancellation terms, component availability, and customer concentration deserve equal scrutiny. A large order can strengthen negotiating leverage or weaken it, depending on whether scarce suppliers and powerful buyers capture most of the economics. A durable winner must turn accelerator demand into recurring storage, networking, services, and support revenue without surrendering the economics to suppliers.
READ NEXT: NVIDIA (NVDA): What Foxconn and Super Micro Are Telling Us about the AI Boom and Pony AI Is Scaling Robotaxis Fast—Can the Stock Reach BofA’s $17 Target?
Hedge Fund Resource Center