As mega-cap technology names prepare to report first-quarter results, the spread between the Cboe Dispersion Index and its three-month implied correlation counterpart has widened to historically extreme levels. NVDA options skew shows a 38% premium for one-month at-the-money implied volatility over six-month, compared to a five-year average of 22%. Market participants have allocated $1.2 billion notional into dispersion trades targeting NVDA, highlighting expectations for outsized stock-specific moves. Should NVIDIA’s quarterly revenue beat consensus of $26.4 billion, individual equity implied volatility is expected to decline by up to 15%, triggering a rapid unwind of existing dispersion positions within two weeks of the earnings release.

In a strategic shift, Chinese authorities have authorized the import of NVIDIA’s newest H200 AI accelerator, marking the first such green light since export controls tightened last summer. Sources at two major mainland data centers indicate they placed orders for over 3,500 H200 modules, representing roughly $1.1 billion in backlog. Analysts estimate this approval could contribute up to 8% incremental revenue to NVIDIA’s Data Center segment in the coming quarter, partially offsetting headwinds from U.S. export restrictions. The decision underscores China’s balancing act between advanced AI needs and domestic semiconductor ambitions, and it reopens a key market that accounted for 16% of NVIDIA’s fiscal-2025 data-center sales.

NVIDIA committed $2 billion to purchase newly issued CoreWeave equity at an $87 per-share equivalent valuation, raising its stake to just over 10% of the AI-focused cloud-GPU provider. The pact includes collaboration to build out 5 gigawatts of AI data-center capacity by 2030, aligning with NVIDIA’s target of supporting hyperscale customers’ multi-year infrastructure road maps. CoreWeave will deploy NVIDIA’s latest Blackwell GPUs and DGX AI systems across new facilities in Texas and Northern Virginia, potentially adding $400 million in incremental GPU module sales annually through 2027. The move both cements NVIDIA’s role as a primary hardware and software partner for AI factories and signals continued locked-in demand from hyperscalers and enterprise users.