GPU / accelerator
GPU and accelerator companies design the chips that train and run AI models. This is the most direct AI compute exposure, but it also carries product-cycle, customer concentration, supply, and valuation risk.
This segment sits inside the physical and semiconductor supply chain that supports AI training and inference. The important question is not only whether AI demand grows, but where scarcity shows up: chips, memory, networking, power, sites, or specialized services.
For 13F analysis, the tag is meant to separate the specific bottleneck a company is exposed to from the broad AI label. That helps distinguish a direct accelerator supplier from a power-equipment company, a data center landlord, or a miner with a power portfolio that could be repurposed for HPC.
- Merchant GPUs and accelerators
- Custom ASIC programs
- Networking and software platforms around accelerators
- Server OEM and ODM integration
- Accelerator generation transitions
- Hyperscaler capex mix
- Gross margin sustainability
- Custom ASIC substitution risk
Situational Awareness LP
Leopold Aschenbrenner · Q4 2025 filed 2026-02-11
The GPU and custom-accelerator vendors themselves (NVIDIA, AMD, etc.) as opposed to suppliers further up the stack.
| Ticker | Name | Value | Instrument |
|---|
| Ticker | Action | Value |
|---|---|---|
| NVDA | Closed | -$299M |