AI Competition's Essence Is Chip Procurement
The 2024-2026 AI competition is an era decided not just by model smartness but by "how much compute you can amass." OpenAI, Anthropic, Google, Meta, xAI secure GPU/TPU at hundreds-of-billions-of-yen scale, and governments advance trillion-yen-scale investment in sovereign-AI strategies.
NVIDIA's Dominance and Challengers
NVIDIA H100 / H200 / B100 / B200 / GB200
The Hopper generation (H100) in 2024, successor H200, and from 2025 the Blackwell generation (B100, B200, GB200) mass-produced. $40-50k per card, one Blackwell GPU rack (GB200 NVL72) on the order of 100M yen. Supply tightness makes 6-12-month waits the norm.
Google TPU v5/v6
Google's in-house TPU updated v5p (2023), v5e (2024), v6 Trillium (2025), v6e (2026). Supports Gemini training/inference. Also provided externally via GCP (Anthropic, various startups).
AWS Trainium / Inferentia
AWS's AI-dedicated chips. Trainium2 (2024) for training, Inferentia2 for inference. Large-scale adoption in exchange for investing in Anthropic. Targets 30-50% cheaper than H100 on cost-performance.
Microsoft Maia / Cobalt
Microsoft's in-house AI chips announced in 2023. For OpenAI-model operation and Azure customers. Aims to reduce NVIDIA dependence and improve Azure margins.
Dedicated Inference ASICs
- Groq LPU: ultra-low-latency inference-dedicated. Llama 3 over 500 tokens/sec
- Cerebras WSE-3: a giant AI chip where 1 chip = wafer-sized
- SambaNova: enterprise-dedicated
- Etched Sohu: a Transformer-dedicated ASIC
China's Situation
US export controls ban H100/H200 export to China. NVIDIA sold regulation-compliant versions (H800, H20, L20, L2) but controls tightened from 2024. China's response:
- Huawei Ascend 910B/910C: performance 60-80% of H100
- Startups like Biren, Moore Threads
- SMIC (China's TSMC equivalent) strengthening 7nm/5nm manufacturing
- DeepSeek etc. proving "frontier models possible even with regulation-compliant versions"
Bottlenecks
| Bottleneck | Status |
|---|---|
| GPU manufacturing | TSMC 5nm/3nm monopoly. CoWoS packaging tight |
| HBM memory | 3-firm oligopoly: SK Hynix, Samsung, Micron |
| Power | Data-center power orders-of-magnitude; nuclear/renewable/self-gen |
| Liquid cooling | Mandatory in Blackwell gen, huge facility-retrofit cost |
| Location | Competition for land suited to transmission/water/cooling |
2026 Highlights
- Blackwell supply stabilization and next-gen "Rubin" announcement
- Full lineup of TPU v6 / v7
- Progress of China SMIC's 5nm mass production
- Expanded production adoption of dedicated ASICs
- Geopolitical impact from power constraints
- OpenAI achieved its US AI-compute-infrastructure 10GW target several years ahead of schedule (April 2026)
- NVIDIA shipped "Vera," its first CPU designed for AI agents, to Anthropic, OpenAI, SpaceXAI, and Oracle Cloud (a dedicated CPU now joins the GPU-centric landscape)
Implications from a Corporate View
- Secure cloud GPU prices with long-term contracts
- Compare inference cost across AWS Trainium, Azure ND H200, Groq, Together AI
- Dedicated ASICs lead for uses needing ultra-low latency
- Self-hosting (vLLM etc.) trade-offs vary by model and GPU price
Summary
AI competition extends to the geopolitics of silicon, power, and cooling. NVIDIA's dominance continues for now, but hyperscaler in-house chips and dedicated ASICs accelerate encroachment. The key for companies is a multi-cloud/multi-chip strategy to contain cost and supply risk.
June 2026 update
- NVIDIA Nemotron 3 Ultra released: Now treated as the strongest U.S.-origin open reasoning model, going head-to-head with Chinese open releases like Kimi K2.6 and MiniMax M3. Chinese models still lead on some academic benchmarks, but the ceiling for U.S. open-weight reasoning models has clearly been pushed up (following the May 22 Cerebras × Kimi K2.6 throughput showdown)
- Vera Rubin AI servers enter mass production: After the May 19 Vera CPU shipments, NVIDIA announced full mass production of the Vera Rubin platform on June 1. The Blackwell → Rubin transition is tracking faster than the original roadmap, putting long-term GPU contracts of frontier labs and hyperscalers up for review
- RTX Spark (GB10) and GB300 Grace Blackwell Ultra expand to PCs and laptops: At Computex 2026 / GTC Taipei, NVIDIA announced an "AI agent PC" push (chasing the estimated $200B CPU market) with Microsoft, Dell, and HP. Grace Blackwell superchips are moving from cloud into clients, putting the layer where local AI agents can realistically run on Windows devices into place
- Physical-AI and self-driving stack disclosed in one event: GTC Taipei also unveiled the "NVIDIA Cosmos 3" world model, a driving brain, an open humanoid robot platform built with Unitree, and a 32B open reasoning model targeted at robotaxis. Beyond GPUs, the "software + hardware + simulator" bundle increases NVIDIA's gravitational pull across the physical-AI stack
- Intel announces Crescent Island GPU at Computex 2026 (up to 480GB VRAM): A single-card play to fit very large LLMs in VRAM. This is the most substantive direct counter to NVIDIA dominance in some time and gives inference-side workloads (where VRAM ceiling matters) a credible alternative to the Blackwell/Rubin path
June 2026 update
- SpaceX leases AI compute to Google for $920M/month per SEC filing (The Decoder). The deal gives Google access to ~110,000 NVIDIA AI chips for Gemini Enterprise demand. That one of the largest cloud providers has to rent compute externally underscores how scarce AI infrastructure has become, and how tightly big-tech business lines are now intertwined. The arrangement also previews a new structural role: non-cloud GPU holders acting as wholesale GPU suppliers inside the NVIDIA economy — notable on the eve of SpaceX's IPO.
- NVIDIA released Nemotron 3.5 ASR (MarkTechPost) — a cache-aware 600M-parameter streaming ASR model that transcribes 40 language locales in real time from a single checkpoint. Less a new research direction than a deployment-focused optimization, it signals NVIDIA pairing its GPU dominance with its own "specialized model × inference optimization" product lineup.
