The rental prices for older NVIDIA GPUs, like the Hopper family and A100s, are increasing. This counterintuitive trend shows demand for AI compute is so far outstripping total supply that even previous-generation hardware is becoming more valuable, highlighting the severity of the GPU crunch.
After addressing GPU and memory supply, NVIDIA is making deals with optical networking firms like Corning. As data center bandwidth needs escalate, high-speed optical components are the next critical performance bottleneck, and NVIDIA is moving to control this layer of the hardware stack.
Analyst Ross Gerber compares the euphoria around the SpaceX IPO to the AOL Time Warner merger of the late 90s. He warns that a wave of mega-IPOs (SpaceX, Anthropic, OpenAI) could put excessive pressure on markets to absorb the new stock, potentially marking the beginning of a bubble.
Analyst Ross Gerber argues a large portion of SpaceX's valuation is tied directly to Elon Musk's leadership, not just business fundamentals. This "Elon premium" creates a massive single-point-of-failure risk for investors, as the company lacks a clear succession plan, making the investment a bet on Musk himself.
To meet surging demand, Anthropic is diversifying its chip supply beyond NVIDIA. An early adopter of Google's TPUs and Amazon's Tranium, its exploration of Microsoft's custom chips reflects a core philosophy of leveraging any available compute resource rather than committing to a single architecture.
NVIDIA now splits data center revenue into "hyperscaler" and "non-hyperscaler" buckets. This strategic reporting change is designed to showcase growth from enterprise and sovereign AI clients—a market where NVIDIA faces less competition from in-house chips and which investors see as a key future growth driver.
The White House's AI regulation approach is shifting due to an internal power struggle. With former AI czar David Sacks's influence diminished, national security voices are gaining ground. This is evidenced by the Office of the National Cyber Director, not a traditional tech office, leading the new executive order.
Unlike general-purpose NVIDIA GPUs, Microsoft's custom Maya 200 chip focuses specifically on running existing AI models (inference). Microsoft claims this makes it cheaper for certain tasks, like its own Copilot tools, creating a cost-saving value proposition for potential customers like Anthropic.